Prediction of Compressive Strength of Mechanical Sand Concrete Based on BP Neural Network
The compressive strength of machine-made sand concrete is influenced by various factors,and it is neces-sary to analyze its strength characteristics in order to improve the quality of concrete.In this paper,a BP neural network with nonlinear characteristics,learning ability,and adaptive ability was used to analyze the compressive strength of ma-chine-made sand concrete.Stone powder content,cement,fly ash,water,machine sand,gravel,and curing age are taken as input parameters,while compressive strength was taken as the output parameter.A BP neural network model with 6 hidden layer nodes is constructed.The simulation results showed an average relative error of 3.47%and a linear correlation coefficient greater than 0.99,indicating that the model has good predictive performance.
BP neural networkmachine-made sand concretecompressive strength prediction
NETL
NSTL
万方数据
