Compressive Strength Prediction of High-Strength Concrete Using Long Short-Term Memory and Machine Learning Algorithms

Compressive strength is an important mechanical property of high-strength concrete (HSC), but testing methods are usually uneconomical, time-consuming, and labor-intensive. To this end, in paper, a long short-term memory (LSTM) model was proposed to predict the HSC compressive using 324 data sets with five input independent variables, namely water, cement, fine aggregate, coarse superplasticizer. The prediction results were compared those conventional support vector regression (SVR) four metrics, root mean square error (RMSE), absolute (MAE), percentage (MAPE), correlation coefficient (R2). showed that accuracy reliability LSTM higher R2 = 0.997, RMSE 0.508, MAE 0.08, MAPE 0.653 evaluation metrics 0.973, 1.595, 0.312, 2.469 SVR model. recommended for pre-estimation under given mix ratio before laboratory compression test. Additionally, Shapley additive explanations (SHAP)-based approach performed analyze relative importance contribution variables output strength.