Exploring the Relationship between NDT and DT Techniques in Concrete: Linear and Non-linear Models

Abstract

The present study involves the development of a correlation of Non-Destructive Testing (NDT) and Destructive Testing (DT) techniques for evaluation of concrete strength. The models selected are Polynomial model (1 to 8 order),Exponential(1 to 2 order), Gaussian( 1 to 2 order), Sum of Sine( 1 to 8 order) and Sigmoidal(Logistic and Gompertz) were the present study. Concrete with grade from M10 to M40 were used for investigation, First the compressive strength is determine by NDT by measuring surface hardness using a rebound hammer. After that same sample were tested by DT using compressive testing machine. There is allot of variation in compressive strength measurement by DT and NDT methods due to various reasons such as mode, environmental, instrument use etc. Correlation is done in order overcome to determine predicated(NDT) and actual value(DT). The main conclusion can be drawn by the analysis showed good correlations in polynomial models, where SSE, the coefficients of determination (R²) and RMSE. The ranged of SSE, R2 and RMSE various from 0.51 to 0.466, 0.96 to 0.973 and 0.05 to 0.0451 respectively. From the various models it is clear that only polynomial Model higher order percentage difference of error had shown decreases in SSE and RSME by 86% to 92% respectively. From the variation it is clear as once increases the order of curve its accuracy increases therefore complex behavior between DT and NDT, Actual V/S Predation required high level of precaution is required. This can be used to prove that the NDT, when calibrated on DT data, allows for accurate estimation of concrete strength with very low intrusion and time investment. The study highlights how advanced mathematical models can facilitate more accurate predictions for concrete strength—which may decisively determine the safety and durability of larger engineering works.