Prediction Method of Structural Static Performance Based on Data Learning

Aiming at the problem of the high cost of establishing a prediction model in the current mechanical structure optimization, a data learning-based structural static performance prediction method is proposed. Taking the cantilever beam as the research object, a finite element simulation model is established to obtain the displacement field data and construct the boundary Condition-displacement field proxy model, the prediction results show that the distribution trend of the displacement field is consistent with the actual, and the maximum displacement relative error is -0.02% and -0.47% when the load is 1000 N and 1600 N, respectively. The size of the uniform force and the concentrated force are discussed in the paper The effect of the action position on the prediction results of the displacement field. The results show that with the increase of the load amplitude, the prediction error increases. Compared with the uniform force, the prediction error under the concentrated force load is larger, and the error at