Method and device for predicting cable force of buckle cable of concrete-filled steel tube arch bridge based on MLP

The invention discloses an MLP-based cable force prediction method for a buckle cable of a concrete-filled steel tube arch bridge. The method comprises the following steps: S1, determining a multi-target response state of the concrete-filled steel tube arch bridge; s2, determining cable force combinations of buckle cables, establishing a concrete-filled steel tube arch bridge finite element model considering the influence of the construction stage and the initial tangent assembly displacement, inputting different cable force combinations of the buckle cables for calculation, extracting corresponding structural responses, taking the structural responses as learning samples, and dividing the learning samples into a training set, a verification set and a test set; s3, preprocessing the learning sample by adopting a scaling normalization method; s4, establishing an MLP deep neural network, judging network stability and prediction precision through evaluation indexes, and determining the MLP deep neural network; a