Structural Optimization Design Of Composite Laminates Based On Evolutionary Computation

In recent years,the application of lightweight technology has become the future development direction of new energy vehicles.The research and development of composite materials provides one of the effective technical solutions for body lightweight.Iterative efficiency has always been a bottleneck in the design and research of body-oriented composite materials.Therefore,a progressive surrogate model structural optimization method was proposed for the structural design of composite body,in order to improve the computational efficiency of optimization iteration and shorten the design cycle on the premise of ensuring the global optimization solution as much as possible.Based on the evolutionary computing algorithm(Kriging surrogate model differential evolution algorithm,KRG-SDE),this paper optimizes the cladding of the composite vehicle hood.The main contents are as follows:(1)Based on the classical theory of single layer plate,the stress relationship of orthotropic composite laminates and the stiffness of composite laminates are analyzed,which lays a theoretical foundation for the subsequent practical application;The finite element analysis program is written,and the accuracy and reliability of the program are verified by cylindrical shell model.(2)By comparing the performance and characteristics of the three basic evolutionary algorithms through the test function,the differential evolution algorithm is selected for improvement;Through three stages of improving model accuracy,model-based optimization accuracy and strengthening global search,the algorithm forms a progressive optimization promotion to achieve a balance between model accuracy and optimization efficiency.Sparse samples were obtained by Latin Hypercube Sampling,and samples were obtained by the Exploration item of EI(Expected Improvement)in the early stage of iteration to improve the model accuracy.With the improvement of model accuracy,in order to accelerate the optimization convergence,Exploitation item was used to obtain better results of objective function in the middle stage of optimization.Considering the error of standard deviation estimation of the surrogate model,in order to further strengthen the global search path,differential evolution is used as the optimization tool when the accuracy of the surrogate model reaches the given threshold.In order to control the computational efficiency of samples,the response of the estimated samples of the surrogate model is used as the supplement of the simulation response at this stage until the optimization convergence.The test function is used to compare the KRG-SDE algorithm with the classical evolutionary algorithm,and the theoretical application performance of KRG-SDE algorithm is verified.(3)KRG-SDE algorithm and DE algorithm were used to optimize composite cylindrical shells,and the optimization ability of KRG-SDE algorithm in engineering application was compared and verified.Furthermore,KRG-SDE algorithm was used to carry out progressive optimization design for the outer plate structure of a steel automobile hood of a company.The overall physical performance of the hood was improved by the multi-stage and equal generation design method,which verified that the proposed algorithm could improve the efficiency of the optimization design of composite laminates.The simulation results show that compared with the traditional steel structure hood,the weight of the hood is reduced by 33.14%and the first mode frequency of the hood is increased by 31.07%.It is verified that the optimization algorithm has obvious advantages in efficiency,robustness,and engineering practicability.