High correlated variables creator machine: Prediction of the compressive strength of concrete

In this paper, we introduce a novel hybrid model for predicting the compressive strength of concrete using Ultrasonic Pulse Velocity (UPV) and Rebound Number (RN). First, collect 516 datasets from 8 studies UPV Hammer (RH) tests. Then, propose High Correlated Variables Creator Machine (HCVCM) to create new variables that have better correlation with output in order improve prediction models. Three single models, including Step-By-Step Regression (SBSR), Gene Expression Programming (GEP), an Adaptive Neuro-Fuzzy Inference System (ANFIS) as well three i.e. HCVCM-SBSR, HCVCM-GEP, HCVCM-ANFIS are employed predict concrete. The statistical parameters error terms such coefficient determination, Root Mean Square Error (RMSE), Normalized (NMSE), fractional bias, maximum positive negative errors, Absolute Percentage (MAPE) computed evaluate results show can than all other HCVCM improves accuracy ANFIS by 5% 10% RMSE, 3% NMSE, 20% MAPE, 7% error.