An Investigation of Genetic Algorithm Based Mechanical Property Optimization Method for Nuclear Piping

In the nuclear steam supply system, the magnitude of nuclear piping is large and their layout are complicated. In order to keep the nuclear piping satisfing the design requirements, the optimization of the location and function of supports on the piping is significant during the design. The traditional method for optimization is manually trial calculation. The limitations of traditional method are high labor costs and experience dependency. Moreover, it is hard to obtain layout scheme with best mechanical property in the traditional method. In present work, a genetic algorithm based intelligent optimization method for piping layout is proposed. The mechanical analysis and genetic algorithm is combined in the present method, and the optimization process becomes automatic. The results show that this method can optimize the location and function of supports effectively. Comparing with traditional method, the present method is more efficiency and more practical.