The Response Surface Method In Structural Optimization Study On The Application

Due to the rapid development of computer technology,the optimization technology has recently emerged rapidly,but when we ask for the optimal value,we usually fit the functional relationship between the factor and the response.This function relationship is mostly in the form of an implicit function.It is more complicated and difficult to express.It is very difficult to ask for this optimal value.The response surface method is a new one in the optimization technology.The combination of statistics and optimization innovatively avoids the complicated solving process of the optimization function in the optimization process.The function.The appearance of the response surface method provides a new idea for the optimization problem.Firstly,according to the principle of response surface method,this paper summarizes the general process of response surface method and makes a careful study of the whole process of response surface method.Response surface method firstly designs the variables,then adds the center point to factor analysis.Then,the first-order model is used to rapidly approximate the optimal neighborhood,and then whether the model is a first-order or second-order model is judged by whether or not there is a bending effect.If a simple first-order model can be fitted,an optimal solution can be obtained directly through a first-order model.If there is no bending effect,a second-order model is generally used to remove factors that have a low effect.Reduce the number of trials,then use the orthogonal design to find out the approximate range,then use BBD center compound or CCD design to find the best neighborhood,and finally find the optimal solution in the best neighborhood and verify.Then it explores the application of the response surface method which is rarely researched in recent years in shape optimization.In the traditional design concept,the shape optimization is very complex,often one point of change,will lead to a series of chain reactions,the amount of calculation is very large,optimization is very difficult.This article has studied this issue,taking the mature structure of a truss structure as a case.Firstly,the physical model was constructed.Using Ansys to simulate can reduce the experimental consumption and simplify the experimental operation.Theapplication of response surface method can effectively reduce the number of trials,and once again using the previous empirical data can simplify the response surface method to find the path,and then The mathematical model of the structure,and then use the professional statistical software Minitab to solve the optimal solution we need the target(constraints in this example is less than 0.5mm displacement,the goal is the minimum quality)and finally compare Minitab and Design Expert optimization results and analysis.As a result of the final optimization,the total truss volume is112.72dm3,which is a reduction of 37.7% compared with the conventional design(displacement0.723 mm,volume 117.90dm3).The mass also decreases by 4.4%.The response surface method not only satisfies our requirements(displacement constraints).,And it reduces the use of materials(target quality),which shows that it can save time and cost in engineering,and has its practical value.Finally,we explored the optimization of dimensions.The chute equipment plays an indispensable role in today's logistics transportation industry and production automation industry.It is a component of automatic sorting and automated transportation center in the logistics industry.Generally,the chutes have specific requirements on the shape.When optimizing the structure,the size optimization is usually used.This paper explores the size optimization of straight chutes using response surface methodology,and determines the proper size range from empirical analysis,force analysis,etc.,to reduce the response surface method search range,thereby reducing the number of experimental iterations,using Simulation in Solidworks.The module conducts iterative optimization of the straight chute model,and then compares it with the Simulation module optimization method using the response surface method and explores the size optimization.The result is obvious.The response surface method not only works better under the same number of iterations.And it is more flexible,unlike the Simulation module optimization method is only a limited option under the optimization,response surface method can be adjusted experimentally at each step.The final optimization result is: mass 64 kg,displacement 1.42e-8m.Compared with the initial data quality of70.21 kg,the displacement is 3e-8m,the quality is reduced by 8.84%,the displacement is reduced by52.67%,and the quality that meets the design requirements is the target mass of 64 kg.This shows that it can save time and costs in engineering and has its practical value.Then,in order to compare with the straight chute,the spiral chute was studied,the spiral chute was empirically and dynamically analyzed,and the spiral chute was optimized by the ordinary method exhaustive method,and the stress of all size combinations was enumerated.For the displacement situation,choose the combination of the lightest fixtures at the stress displacement limit.Comparing the response surface method with the exhaustive method,it is found that when the number of independent variables islarge and the relationship between the response and the independent variables is complex,and there are experimental values,the response surface method is more appropriate.