Optimization Of The Robustness Of Multi-hole Columnar Compressive Structure

The pressure structure of manned submersible provides space for the arrangement and protection of numerous equipment and personnel,and it is an important guarantee for its safety and efficiency underwater operation.In the process of design,construction and use of pressure structure,there will inevitably be some uncertainty factors,the existence of these uncertainties will lead to the performance of pressure structure fluctuations,and further affect the safety and efficiency of manned submersible.In addition,the pressure structure of manned submersible is a typical pressure structure with multiple holes,which will not only affect the performance of the pressure structure,but also lead to the further complexity of its design and research.Therefore,it is of great theoretical significance and engineering reference value to carry out the study of uncertainty optimization for multi-hole pressure structure of manned submersible.For this reason,this paper carries out uncertainty optimization research on multi-hole columnar pressure structure,and the specific work is as follows:Firstly,according to the specification,carry out the initial design and calculaing check of the pressure structure without holes and complete the determination of the pressure structure’s opening and strengthening form.Through finite element software ABAQUS,the multi-hole columnar pressure structure was analyzed by finite element method,and its strength and stability were checked,and the initial design scheme was obtained.Secondly,to complete the sample points to obtain,with the help of the Python language,carries on the secondary development of ABAQUS,through the ISIGHT software for ABAQUS software integration,solve the communication problem between with ISIGHT and ABAQUS,and further realize the model is established,the load applied,boundary conditions are set up and mesh generation automatically of pressure structure,finally realizes the pressure structure parametric design and analysis.Thirdly,based on the parametric analysis process of pressure structure,the influence of design variables on the response is analyzed,and the variables with less influence are selected to achieve the purpose of reducing the dimension of design variables and reducing the difficulty of problem research.The sample points were further selected for parametric analysis,the approximate model was fitted for the sample points,the approximate model technology with the minimum error of corresponding response was selected to complete the establishment of the highprecision approximate model,and the accuracy of the approximate model was further checked by using the finite element analysis method.Forth,introduce the single objective optimization and multi-objective optimization method,respectively adopt the method of ASA,MIGA and NSGA-Ⅱ,carry out design of pressure structure for single objective and multi-objective optimization,and compare and analysis the advantages and disadvantages of three kinds of method.Finally,the robust optimization of 6 quality level is introduced.Using method of Monte Carlo sampling to analyze the robustness of the deterministic optimization,pressure structure respectively for single response and multiple responses multiobjective robustness optimization,The two robust optimization methods are solved and studied with different iteration times and different search ways,the results of a variety of schemes for adequate comparison and analysis,finally the scheme with the highest robustness is obtained.