Extraction of optimum ratio parameters of steel fiber reinforced concrete based on BP neural network

Abstract

In order to overcome the problems of low precision in the traditional extraction method of steel fiber reinforced concrete optimum ratio parameters, a new extraction method of steel fiber reinforced concrete optimum ratio parameters based on BP neural network is proposed in this paper. Based on the analysis of suspension state, virtual accumulation state and actual accumulation state of concrete solid particles, combined with the characteristic particle sizes of large particles and small particles, the packing density and constraint conditions of concrete solid particles are calculated. Based on the structure of BP neural network, the input of hidden layer is calculated, and the output of hidden layer is calculated according to s function. By comparing the expected output with the actual output, the threshold and weight of BP neural network are modified to construct the optimal ratio parameters of steel fiber reinforced concrete. The experimental results show that under the same experimental environment, the flexural strength of concrete is the best under the ratio parameters extracted by this method, and the extraction accuracy of the best ratio parameters is the highest, so this method can improve the rationality of the ratio of steel fiber reinforced concrete.