Optimization Technology in CAE Method (2)

Therefore, the software has a unified database is the premise of the efficient CAE optimization process. This unified means that the data used before and after processing and the solution should be in the same database, rather than passed through the data file, which is bound to reduce optimization. The efficiency of the process. In addition, most pre-processing and solvers of software that pass data through files are not fully supported. Pre-processed data files often need to be manually modified before being put into the solver. This is incompatible with the automation of the optimization process. Once this happens and cannot be avoided, either give up or program the data file automatically.

Third, the optimization process is actually a process of continuously correcting design parameters automatically. Therefore, in order to ensure the smoothness of the optimization process, CAE software must have complete and efficient parameter flow control technology. In the process control process, not only the design data to be optimized can be parameterized, but also the flow control is required to have the ability to judge branches and loops so that the software can automatically cope with various complicated situations that arise in the optimization process of large problems.

Fourth, high-precision grids are one of the key factors for successful finite element analysis. A good CAE software to deal with the optimization process, especially the shape optimization problem, must have an intelligent mesher to solve the mesh singularity problem that occurs when the shape parameters change sharply. Fifth, modern CAE software usually has and should have nonlinear processing capabilities, and the convergence control of nonlinear problems has caused countless heroes to bend. Generally, the means to improve the convergence of nonlinear problems should be determined on a case-by-case basis, and the optimization process for a nonlinear problem tends to converge due to various factors. However, the optimization process is an automatic control of the iterative process, and people cannot participate too much. Therefore, the nonlinear convergence intelligent control technology is indispensable for the nonlinear optimization problem.

When it comes to nonlinearity, one might think of a solution technique called explicit integration. This technique is often used to solve high-speed deformation and highly nonlinear problems, complementing the implicit solution techniques commonly used to solve static or slow dynamic problems. Most of the problems can be solved by choosing only the right one, but not all problems can be separated, such as stamping and rebound process simulation. Usually, the stamping process is simulated in an explicit way, and the rebound is simulated in an implicit manner. Process, then there must be an explicit to implicit switching process here. If you simply simulate these two processes, this kind of switching can be done manually. However, for people who do not participate in the optimization process, if the switching cannot be performed automatically, the optimization analysis of such problems can not be completed. When the software application level reaches a certain height, people may think of trying a cooperative optimization method, that is, multiple networked work machines of the same work group work together to optimize the same problem. Usually the model brand or even the operating system of each working machine in the same working group may be different, then the incompatibility problem of the database of different platforms may make such a creative attempt become a bubble.

Of course, not all software has this problem. Today's popular CAE software, ANSYS, has a strong influence on this issue, and with some of her other characteristics, she has become a worthy role in the current topic. ANSYS is a CAE software that combines structural, thermal, electromagnetic, and fluid analysis capabilities. It can perform multi-field coupling analysis. She has strong pre- and post-processing capabilities, especially in smart mesh dividers; she has Strong explicit or implicit nonlinear solution ability, and explicit and implicit can be automatically switched automatically; nonlinear convergence control is intelligent, and the convergence of nonlinear problems can be completed without manual intervention for most engineering problems; She also has a parametric design language ━APDL that is praised by its users as “all-powerful”. The language has high-level language elements such as parameters, mathematical functions, macros (sub-processes), judgment branches and loops, and is an ideal program flow. Control language; her pre- and post-processing and database uniformity and the compatibility of different platform databases make her suitable for advanced optimization analysis.

IV. An example of CAE optimization The turbine unit of Ertan Power Station is the largest single unit capacity unit used so far in China, with a volute diameter of 20 meters. It consists of two parts: one part is a seat ring structure welded by two annular plates and 20 fixed guide vanes, which constitutes the bracket of the housing part; the other part is the shell which is welded end by end of the 25-section cone. Body structure. Each cone has a different diameter and a different thickness. At the same time, each segment is not a complete cone, but a part is cut along the axial direction. The axial straight edge of the remaining portion is circumferentially welded to a position of the upper and lower ring plates. The choice plays a key role in the stress distribution throughout the volute. If this position is chosen well, the stress distribution in the volute tends to be uniform, and the maximum stress in the entire structure is reduced, so that the thickness of the volute casing and the upper and lower ring plates of the seat ring can be reduced to a certain extent to reduce the thickness thereof. Volume or weight to reduce material, processing, transportation and installation costs. Therefore, the purpose of this analysis is to select the shell and the upper and lower sides under the premise of ensuring that the water flow characteristics of the flow channel and the maximum stress of the entire structure do not exceed the allowable stress (the equivalent stress value of the entire structure that the manufacturer provides should not exceed 160 MPa). The welding position of the seat ring and the thickness of the seat ring plate and the housing material make the entire structure the lightest. The optimal parameters of the volute underground structure shape (18 in total) are 13 shell thicknesses, 1 ring plate thickness, 4 welded joint positions (other positions are linear interpolation of the four points), which is a design variable. Numerous fluid-structure coupling optimization problems, as well as the water conservancy characteristics of the flow channel, that is, the cross-sectional area of ​​the flow channel must not be reduced. The modeling process makes full use of the parametric modeling function of ANSYS, and uses the APDL language to establish the parametric model of this problem. The ring plate and the shell wall were divided by SHELL63 (shell unit), the guide vane was divided by SOLID45, and the fluid model was split by Fluid142 unit. The objective function of the optimization of this problem is the total volume of the structure. By comparing the data before and after the optimization, the total volume of the structure is reduced from 56.84 before optimization to 52.26, and the reduction is 8% of the original design. In addition, after optimization, the water pressure of the shell wall tends to be smooth, and the stress distribution inside the structure tends to be uniform.

V. CAE Optimization Method in Development With the development of CAE technology, the optimization technology in CAE will continue to develop. In addition to the above-mentioned features will become more and more obvious, there will be many new features. Such as discrete optimization problems, multi-objective optimization problems, etc. A new method called topology optimization has emerged in modern CAE optimization technology, which has been adopted to some extent by some CAE software (including ANSYS). With the gradual maturity of its theoretical foundation, the practicality will gradually improve. It is believed that topology optimization will be a good complement to the classical optimization method.

Previous page

Bar tools

Wine rack,stainless steel wine rack,Iron wine rack, wine holder,Goblet Holder