Applied Research On Response Surface Method In Structural Optimization

Lightweight materials can be used in product lightweight research,but their costs are often high,to meet the certain strength and stiffness,improve the design quality in order to achieve lightweight design for the early design to reduce costs.In order to improve the design model and design parameters,the main goal of this paper is to compare the predecessors' examples to reflect the advantages of the response surface optimization method: global search,the number of iterations is small,can reduce the volume or weight The Using the function relation to fit the actual simulation model,the screening experiment is used to determine the optimization direction.The main work is to find the best field and get the optimal solution.The essence of the response surface method is to reduce the test on the basis of more accurate fitting The target value,reduce the uncertainty error ? value,try to balance the number of tests and the effect,in order to achieve high efficiency balance point,so that the design effect is the best.RSM is a sequential method,the first factor analysis to determine whether the first-order linear regression model or the second-order surface model,if the variance analysis of all P values are less than 0.05 for the second-order model,and then adjust the center and horizontal range search excellent.The fitting equation of RSM is nothing more than a first order model or a second order equation.The first-order model can use the least squares method to locate the optimal solution quickly.The second-order model first finds the best neighborhood and then finds the optimal solution in the best field.The optimal condition of the optimal neighborhood is the actual constraint S and the fitting constraint the relative error of ?₀ is less than 5%.The optimal solution is that the fitting constraint is smaller than the actual constraint and the volume is minimum.The object of the study is the truss bar,the crank and the triangular arm,the computer embedded in the APDL program using the computer CAD and CAE technology and the stress and volume can be obtained.Using Soildworks modeling analysis,the optimization software Minitab and Design-expert are used to integrate Optimization,must be required to have the basic knowledge and experience of the project,and ultimately to solve and optimize the actual structure of the project,through the response to the surface of the feedback adjustment,within the global,artificial set the center and the search level,effectiveness.In the example of truss bar,it is necessary to carry out co-optimization,that is,considering the shape change and size change under the same time to optimize the two,the difficulty lies in the shape change and the size of the mutual coupling,using the response surface method not only without Lagrange multiplier,But also to avoid the shape of the change and the size of the coupling between the changes,by adjusting the test center point and the search level range,in response to the direction of the optimization.In the optimization process,it is found that the truss rod 5 has a coupling,and the expression is stress concentration.Therefore,it is necessary to increase the size of A5 cross section.These are the artificial control of the experimenter.According to the feedback method of the response surface method,But also to respond to the efficient source of the surface method,only the iterative three times to find the shape of the best,iterative two times to find the size of the optimal point,the last iteration once to find the optimal solution for collaborative optimization design.In the case of connecting rod and triangular arm,we use Soildworks to model and optimize with its own algorithm,but the optimization precision is not high.Therefore,we use the response surface method to optimize the quadratic optimization and find the optimal solution in the best field.