Prediction of the compressive strength of high-performance self-compacting concrete by an ultrasonic-rebound method based on a GA-BP neural network

Abstract

To address the problem of low accuracy and poor robustness of in situ testing of the compressive strength of high-performance self-compacting concrete (SCC), a genetic algorithm (GA)-optimized backpropagation neural network (BPNN) model was established to predict the compressive strength of SCC. Experiments based on two concrete nondestructive testing methods, i.e., ultrasonic pulse velocity and Schmidt rebound hammer, were designed and test sample data were obtained. A neural network topology with two input nodes, 19 hidden nodes, and one output node was constructed, and the initial weights and thresholds of the resulting traditional BPNN model were optimized using GA. The results showed a correlation coefficient of 0.967 between the values predicted by the established BPNN model and the test values, with an RMSE of 3.703, compared to a correlation coefficient of 0.979 between the values predicted by the GA-optimized BPNN model and the test values, with an RMSE of 2.972. The excellent agreement between the predicted and test values demonstrates the model can accurately predict the compressive strength of SCC and hence reduce the cost and time for SCC compressive strength testing.

Hammer rebound index as an overall-mechanical-quality indicator of self-compacting concrete containing recycled concrete aggregate

“…The hammer rebound index and the compressive strength relationship is therefore modified, which in turn means that conventional formulas are invalid for SCC [27]. Nevertheless, some research has been focused on the extent to which varying proportions of fine aggregate and flowability in the fresh state can affect the hammer rebound index of SCC [28], from which models have been developed that relate it to compressive strength [29].…”

Prediction of the shear capacity of ultrahigh-performance concrete beams using neural network and genetic algorithm

“…Data standardization processing is the key step of neural network technology and can greatly improve the learning efficiency of ANNs. It can eliminate the difference in dimensions between different data 66 , 67 and control the data range within [0,1]. The standardized processing formula is shown in Eq.…”

Study on the Residence Time and Texture Prediction of Pea Protein Extrusion Based on Image Analysis

“…Currently, the optimization of neural networks is primarily achieved to attain the optimal prediction model. Particle Swarm Optimization (PSO) and Genetic Algorithm (GA) are common model optimization methods [16] which are both capable of ameliorating situations in which BP neural networks can easily fall into local optima, thus achieving optimal prediction effects [17,18].…”