Study on Axial Compressive Load Bearing Capacity of Concrete-Filled Double-Skin Steel Tubular Columns Based on Particle Swarm Optimization BP Neural Network
Circular concrete-filled double-skin steel tubular(CFDST)columns have become the main bearing components in modern engineering structures due to its unique structural form and excellent mechanical properties.However,the mutual restraining effect between the outer steel tube,inner steel tube and core concrete leads to the complexity of its stress.For this reason,in this paper,the axial compressive load bearing capacity of circular CFDST columns is investigated using PSO-BP hybrid neural network algorithm.167 sets of data were collected to establish a database,8 influencing factors were selected as input layer parameters and axial compressive load bearing capacity as output layer parameters,the defects existing in the traditional BP neural network model were analyzed,and the PSO-BP neural network model was established.In addition,the results of machine learning model is compared with that of three specifications,and the results show that the results of machine learning model has higher accuracy than that of the three specifications.Among them,the PSO-BP neural network model has better prediction ability compared with the BP neural network model,which is more helpful for predicting the axial compressive load bearing capacity of CFDST columns,which is of great significance for the engineering study of the mechanical properties of CFDST columns.
BP neural networkparticle swarm optimization algorithmconcrete-filled double-skin steel tubular columnaxial compressive load bearing capacitymachine learning model
万方数据
维普
