abstract |
The invention provides a method for predicting the compressive strength of novel lightweight concrete based on RBF neural network, comprising the following steps: preparing novel lightweight concrete based on cementitious materials and additives, and obtaining the basic proportioning of the novel lightweight concrete; Describe the basic ratio and single-factor control method of the new lightweight concrete, obtain multiple groups of new lightweight concrete, and obtain multiple groups of sample parameters of the new lightweight concrete; build an RBF neural network; based on the sample parameters, the RBF neural network The network is trained, and the compressive strength prediction of the new lightweight concrete is completed based on the trained RBF neural network. Using the prediction method of the present invention not only can greatly reduce the actual workload, but also has important academic and engineering significance for the lightweight concrete raw material, design and experimental scheme. |