Since the CAD/CAM technology of automotive panels is relatively mature and is not the most critical factor affecting the stamping process, the following focuses on the CAE technology of stamping and its integration with CAD/CAM technology.
1. Stamping forming CAE pretreatment system
The most obvious manual intervention in the computer simulation of the stamping forming process is the pre-processing process. Most of the current stamping simulations use common finite element analysis pre-processing software. Although these pre-processing softwares are very powerful, they are inconvenient to use because they are not considered for the characteristics of sheet metal forming. This is especially true for mold designers who are not familiar with the finite element method. Recently, ETA in the United States introduced eta/DYNAFORM, which takes into account the characteristics of sheet metal forming. However, eta/DYNAFORM is directly served by LS-DYNABD, and the characteristics of sheet metal stamping are not fully utilized. The stamping and forming simulation pre-processing software system recently designed by the Automotive Technology Laboratory of Hunan University has taken into account the characteristics of stamping and forming of automobile panels to the greatest extent, and it is specially designed for the stamping and forming special software developed by the laboratory itself. The software interface is fully considered. The general terminology and concept of the mold designer also has a certain database of mold and process design. Although this system needs to be further improved, it has shown great practicality and advancement.
Car cover stamping and forming CAE pre-treatment systems generally have at least the following functions:
1) Accurately describe the geometry of the mold. According to the requirements of different simulation software, the surface of the mold can be represented by analytical methods, and can also be represented by a collection of finite elements. When the latter is used, the geometry represented by the finite element mesh must be sufficiently close to the designed die surface shape, and this approximation accuracy should be arbitrarily adjustable;
2) Form an arbitrary grid for a shape sheet. The grid unit is generally a three-node triangular unit, or a four-node quadrilateral unit. It should have the function of generating an arbitrary dense mesh, and the mesh area and the dense mesh area should be able to automatically transition. Grid quality should be automatically detected and controlled;
3) Automatic definition of the contact friction interface between the mold and the sheet material and arbitrary characteristic parameter selection;
4) Automatic definition of the arbitrary motion characteristics of the punch;
5) Automatic definition of the rolling rib and the blanking force on the binder ring;
6) Automatic definition of sheet property parameters;
7) The function of arbitrarily modifying the model;
8) The function of graphically displaying all the information contained in the model;
9) Friendly user interface and help function.
2. Stamping and forming CAE software system
Stamping and forming simulation software involves the following key technologies:
1) Geometric and mechanical models of the mold. The rigidity of the mold is much larger than that of the sheet material, so it can be handled as a rigid body in general. However, when it comes to the friction and wear of the mold, the mold should also be considered as an elastoplastic body. The geometry of the mold can be described by a finite element mesh or by an analytical surface/CAD surface. The latter has high geometric accuracy but requires special contact interface processing algorithms. The motion of the punch is usually described by a given displacement history or speed history. The binder ring is usually only allowed to move in the direction of the punch, and the degree of freedom of the die is completely constrained.
2) Sheet deformation mode and housing unit. At present, almost all of the stamping and forming simulation software of the panel assumes that the deformation mode of the sheet material conforms to a certain shell theory, that is, the influence of the sheet thickness stress on the forming process is neglected. However, this assumption is often not true. The stress of the rib portion or the region with a relatively small bending radius is closer to the three-dimensional state. The influence of this local three-dimensional stress state on the simulation results of the forming process has begun to attract the attention of researchers. The housing unit is most commonly used in low-order bilinear units (including triangular and quadrilateral elements). This is not only because low-order units facilitate the processing of computational and contact interfaces, but also because they are best suited for simulation algorithms.
3) Sheet constitutive model. The sheet material used for stamping of automobile panels has obvious orthotropic anisotropy due to work hardening or the like during rolling. It is therefore important to consider this anisotropy when constructing the criteria and flow criteria for the sheet. At the same time, it should be noted that the stamping process itself will change the anisotropy of the sheet. The Barlat and Hill models and their variants are currently used the most. For the simulation of the sheet metal forming process, the computational problems in the constitutive model are also critical. The stress state of the plastic deformation zone must not only satisfy the yield criterion. It is also necessary to satisfy the assumption that the thick-direction stress is zero in the plate-shell theory, which increases the difficulty in calculating the plastic deformation of the sheet. Research shows that the reflection algorithm in the plane stress state has the advantages of high precision and small computational workload.
4) Contact friction theory and algorithm. The stamping of the automobile cover is completely accomplished by the contact force and friction acting on the sheet. Therefore, the calculation accuracy of contact force and friction directly affects the calculation accuracy of sheet deformation, and the calculation of contact force and friction force first requires calculation of the actual contact surface at any given time. This is the so-called contact search problem. In the finite element method, calculating the contact surface is actually finding all the finite element nodes in contact state. Although contact search is essentially a process of geometric calculations, it has important mechanical implications.
There are two basic methods for calculating the contact force, namely the penalty function method and the Lagrange multiplier method. The penalty function method is an approximation method that allows the boundary of contact with each other to generate a penetration and associates the magnitude of the contact force with the amount of boundary penetration by a penalty factor. This method is relatively simple, and is also suitable for explicit algorithms, but it affects the critical time step in the explicit algorithm, while in the implicit algorithm it affects the reversibility of the coefficient matrix in the computer. The quality of the penalty factor also affects the reliability of the calculation results. The Lagrangian multiplier method does not allow the mutual penetration of contact boundaries. It is an accurate contact force algorithm, but it is incompatible with explicit algorithms and requires special numerical processing. The defensive node method is such a processing method.
The calculation of the friction first requires the selection of a friction law that is suitable for the frictional properties of the two contact interfaces. The most widely used is the traditional Coulomb friction law. However, the law has the assumption of a purely adherent state, making the explicit algorithm difficult. To overcome this difficulty, either use the penalty function method or use the defensive node method to calculate the friction in the purely adhered state. In the implicit algorithm, the friction slip state will result in an asymmetric coefficient matrix, which increases the difficulty of solving. In recent years, some scholars have proposed the so-called nonlinear friction law based on sufficient experimental observations, thus eliminating the assumption of pure adhesion state in the traditional friction law, which provides convenience for explicit algorithms. However, the surface stiffness coefficient used in the nonlinear friction law is carefully selected according to the physical and chemical properties of the two contact surfaces, and there is not enough experimental data for reference. The more general motor control law uses the elastoplastic theory to define a friction criterion similar to the yield surface and a similar flow criterion to assist the friction slip criterion, and can consider the anisotropy of the friction surface.
5) Grid subdivision and grid adaptive technology. The simulation of the stamping process of a large cover typically involves tens of thousands of finite elements. In order to properly lay out the density of the grid at different stages of stamping, it is necessary to mesh the mesh or mesh adaptive technology of the sheet. Mesh subdivision refers to the division of a unit subjected to a high strain or stress gradient into a number of small units based on a reference grid, while the other units remain unchanged. Grid adaptation means that the mesh is continuously re-divided with the deformation of the sheet to ensure that the high strain gradient region has a dense mesh and the low strain gradient region has a thin mesh. One of the key points in mesh subdivision and mesh adaptive technology is the mutual conversion of physical quantities such as stress and strain at the integration points between old and new finite elements. If this conversion relationship is not handled well, it may bring errors to the simulation results and even invalidate the entire simulation results.
6) Implicit algorithms and explicit formulas. The problem of numerical integration methods in the time domain is involved in the calculation of the stamping process as a dynamic problem. Before the mid-1980s, people basically used the Niuman method to integrate time domain.
The displacement, velocity and acceleration at any time in the bullman method are related to each other, which makes the solution of the equation of motion into a series of interrelated nonlinear equations. This solution process must be implemented by iterating and solving simultaneous equations. This is known as the implicit solution. The implicit solution method may encounter two problems. One is that the iterative process does not necessarily converge, and the other is that the simultaneous equations may be ill-conditioned without a definite solution. The biggest advantage of the implicit solution method is that by setting the appropriate β and γ values, it has unconditional stability, ie the time step can be arbitrarily large.
Due to the convergence of implicit algorithms, the central difference method was used more and more in the mid-1980s for time domain integration of stamping process simulation. In the central difference method, the relationship between displacement, velocity, and acceleration is as follows:
Ui+1=2ui-ui-1+αi(Δt)2(3)
Ui-1=ui+1-ui-1/(2Δt)(4)
It can be seen from the above equation that the displacement at the current moment is only related to the acceleration and displacement at the previous moment. This means that the displacement solution at the current moment does not require an iterative process. In addition, as long as the mass matrix and the damping matrix in the equation of motion are diagonalized, the acceleration solution at the previous moment does not need to solve the simultaneous equations, which greatly simplifies the problem. This is called the explicit solution method. The advantage of the explicit solution method is that it has neither convergence nor solution of simultaneous equations. The disadvantage is that the time step is limited by numerical stability and cannot exceed the critical time step of the system. Since the stamping process has a strong nonlinearity, the time step cannot be too large considering the accuracy of the solution, which largely compensates for the shortcomings of the explicit solution method.
The obvious shortcomings of the explicit algorithm are reflected in the calculation of the rebound. In order to accurately calculate the springback of the stamping part, the calculation of the stamping forming process must be completed after the dynamic response of the workpiece is small enough. The response time is usually several hundred milliseconds or even more than one second, and the explicit algorithm is used to solve the problem. The response of a long process involves a huge amount of computational effort. On the other hand, the nonlinear component in the dynamic response of the workpiece is obviously weakened after unloading. These factors combine to provide the conditions for the successful application of the implicit algorithm in the calculation of the unloading process, and use the explicit algorithm to solve the loading process of the stamping forming. Solving the unloading process with an implicit algorithm is an effective way to comprehensively utilize these two algorithms through repeated practice in recent years.
3. Stamping forming CAE aftertreatment system
The successful application of stamping CAE technology is inseparable from a good aftertreatment system. Text, graphics, and curves are an indispensable expression for post-processing systems. Convenient display of stress, strain, displacement, and workpiece thickness distribution is a fundamental function of the aftertreatment system. The display mode should include not only cloud maps, contours, but also time history curves and numerical outputs of local quantities. In addition to directly displaying the direct results of the CAE software system, the post-processing system must be able to perform engineering calculations and synthesis on these results, thereby obtaining and visually displaying other engineering data such as equivalent stress, maximum stress, etc. From the perspective of forming process analysis, the calculation and display of the following results are particularly important: 1 calculation of forming force, calculation of 2 rebound force, explicit display of 3 wrinkles, display of 4 possible fracture zones, and strain state of 5 different cells The location on the forming limit net.
4. Engineer-assisted process analysis software system
The effective use of stamping and forming CAE systems usually requires a strong background knowledge of finite element methods. At present, most of the pre- and post-processing systems for stamping and forming CAE are general-purpose software, so the degree of specialization is low, which is not conducive to mold engineers, especially to non- Linear finite element methods are used by unfamiliar engineers. In order to overcome this shortcoming, a stamping process analysis software system for engineers is also being developed at home and abroad. This system is based on a stamped CAE software system and its front and rear processing systems, equipped with a dedicated stamping technology database and a good user interface. This kind of system talks with the mold engineering with the terminology, expression and thinking mode of the mold, and the operation is simple and the result is intuitive. It helps mold engineers to determine the shape and size of the blank, the layout of the rolled ribs, the lubrication scheme between the mold and the stampings, the arrangement of the process accessories and process holes, the size of the blanking force, and the adjustment of the die structure dimensions such as the fillet radius. Wait. Of course, every process plan and process parameter adjustment involves a complete stamping simulation. This means that after each process plan and parameter adjustment, the mold engineer has to wait for a while to let the computer "think" about the adjustments made. The "thinking" time of this computer is actually the time required to perform the simulation, depending on the speed of the computer and the size of the simulation model involved, which may be minutes or hours or even tens of hours.
5. Standardized experimental device and inverse calculation technology
The simulation results of stamping are based on the original parameters input to the simulation system, including the constitutive property parameters of the material, and the friction characteristics between the sheet and the mold. The correctness of these parameters naturally directly affects the validity of the simulation results. With the maturity of simulation technology, the acquisition technology and devices for simulation using raw data have attracted more and more attention. The core of this technology is to design a series of standardized experimental schemes to make the sheet produce deformation and stress states similar to those in stamping, and then measure some macroscopic quantities such as displacement and force that are easy to measure accurately, and then use the finite element method to calculate some A hypothetical unknown microscopic amount such as hardening modulus. This experimental technique is important for at least two reasons. First, a large amount of material data was obtained with relatively backward means several years ago. Moreover, material composition and characteristics fluctuate greatly now. It is not enough to obtain raw data from the design manual alone, and the data provided by the material manufacturer is limited in quantity and precision. Secondly, the data obtained by the inverse calculation has better matching with the simulation algorithm.
In addition to simulating the experimental setup with raw parameters, the experimental verification techniques of the simulation software are also important. The reliability of the simulation results is also affected by a series of human factors, such as the reliability of programming, the rationality of model establishment, and the correctness of parameter input, on the premise that the principle and method are completely correct. Therefore, before using a simulation software, especially when developing simulation software, it is necessary to adopt a series of well-designed experimental measures to ensure the correctness of each link. Only when all aspects of the simulation software and the software are verified to be reliable, can there be reason to believe the reliability of the simulation results.
6. Interface and integration technology with CAD/CAM systems
Since the CAD/CAM technology of the mold has become quite popular, the integration of the stamping and forming CAE system and the CAD/CAM system has become an irresistible trend. The key issue here is how the CAE system obtains the geometry and dimensions of the mold from the CAD/CAM system and how the CAE system feeds back the mold design modifications from the simulation results to the CAD/CAM system. The most commonly used graphics data transfer method is to use IGES or VDA format files. Most industrial CAD/CAM systems provide such output files, and CAE systems simply extract and interpret graphical information in the format specified by IGES or VDA.
H7 Led Headlight,H7 Bulb,H7 Headlight Bulb,H7 Led Lights
CHANGZHOU CLD AUTO ELECTRICAL CO.,LTD , https://www.cld-led.com