摘要：As the idea of simulated annealing (SA) is introduced into the fitness function, an improved genetic algorithm (GA) is proposed to perform the optimal design of a pressure vessel which aims to attain the minimum weight under burst pressure con- straint. The actual burst pressure is calculated using the arc-length and restart analysis in finite element analysis (FEA). A penalty function in the fitness function is proposed to deal with the constrained problem. The effects of the population size and the number of generations in the GA on the weight and burst pressure of the vessel are explored. The optimization results using the proposed GA are also compared with those using the simple GA and the conventional Monte Carlo method.

摘要：This paper proposes a theoretical method using finite element analysis(FEA) to calculate the plastic collapse loads of pressure vessels under internal pressure,and compares the analytical methods according to three criteria stated in the ASME Boiler pressure vessel Code. First,a finite element technique using the arc-length algorithm and the restart analysis is developed to conduct the plastic collapse analysis of vessels,which includes the material and geometry non-linear properties of vessels. Second,as the mechanical properties of vessels are assumed to be elastic-perfectly plastic,the limit load analysis is performed by em-ploying the Newton-Raphson algorithm,while the limit pressure of vessels is obtained by the twice-elastic-slope method and the tangent intersection method respectively to avoid excessive deformation. Finally,the elastic stress analysis under working pressure is conducted and the stress strength of vessels is checked by sorting the stress results. The results are compared with those obtained by experiments and other existing models. This work provides a reference for the selection of the failure criteria and the calculation of the plastic collapse load.