The Research On The Method Of Intelligentized Structural Optimization In Culvert

In order to get the best performance in certain aspect of one project, the optimum design was applied in the engineering, but the traditional optimization method needs a great deal of calculation, in this paper, the neural network and genetic arithmetic were combined to implement the optimum design, and find a intelligentized way to analysis and compute the structure.Based on the design data of water transmission culvert in tianjin, in this paper, diverse optimization models were created according to different load forms and work cases, and calculated using genetic arithmetic with design norm constrain condition, the available result data as the patterns to train the neural network, and then the nonlinear relationship of input to output is build up, by using this relationship, we can easily get the optimum result. So we get a intelligentized and simplified method by using the trained neural network.It can be conclude from this paper, this method of optimum design based on the neural network combine with genetic arithmetic used in the field of engineering structural optimization is feasible, can simplified the work. So the method and the conclusions in this paper can find their ways into practical project and research field, and is worthy of being studied and popularized deeply.