Cannon Gun Tube Optimization Method And Application Research On Its Own

To adapt comprehensive performance requirements of barrel under the condition of the modern war, the self-propelled artillery’s barrel is optimized, considering strength, stiffness, weight and other factors. The task in this paper is a part of the twelfth five-year national defense’s preliminary research project. Using modern intelligent optimization algorithms, with the finite element analysis software as the tool, the procedure is written by MATLAB, and typical self-propelled artillery’s barrel taken as a model, the optimization methods are studied, considering multiple disciplinary models influencing barrel performance. The specific completed tasks are as follows.(1)Multiple disciplinary models influencing barrel performance are analyzed, including the interior ballistic model, the structural model, the strength model and the stiffness model.(2)Using the parameter sensitivity analysis method based on the stochastic finite element, design parameters influencing the structural strength reliability are got, which can provide the reference for the selection of structural optimization variables.(3)The single-objective optimization model of barrel is established based on ANSYS, and constraints of the structure, strength and stiffness met, the single-objective optimization design is proceeded with, on the basis of weight of barrel as the objective function.(4)To avoid too many finite element iterations of the optimization, an efficient multi-objective optimization method is proposed, based on weights optimized. Through the uniform test, the multi-objective optimization model which combines neural networks and genetic algorithms is established, using the fuzzy solution to multi-objective optimization. According to training data of test samples, weights are got by weights optimized, and the optimized model of weights is built to overcome the subjectivity of weights determination of traditional methods, based on neural networks and genetic algorithms. On this basis, barrel is optimized by neural networks and genetic algorithms. The suggested method has an improved optimizing effect and a higher efficiency.(5)Multi-objective optimization software program of barrel is developed. Use EXCEL to develop the user interface, and the optimization program is worked out by MATLAB. Then, realize data exchange through the interface technology.