Innovative image analysis-based methods for the estimation of conventional concrete mixture proportions from hardened concrete

Abstract

The importance of concrete mix design is not hidden from anyone. Therefore, extracting mixture proportions from hardened concrete can be considered as a brilliant tool for various purposes such as controlling the accuracy of construction and evaluating the reasons for the inappropriate performance of concrete. For this purpose, an innovative and comprehensive method was developed in which images of the macrostructure and microstructure of hardened concrete were utilized. At first, the volume ratios of aggregates, cement paste, and air voids were estimated using the results of analyzing microstructure images containing images of black and white treated concrete cross sections scanned with a resolution of 2400 dpi and Backscattered electron microscopy (BSE) images at 500 × magnification. Subsequently, with regard to the aggregates volume ratio and images of the macrostructure of concrete, scanned with a resolution of 600 dpi, volume ratios of coarser aggregates (retained on the 0.3 mm sieve) and finer aggregates (passing the 0.3 mm sieve) were calculated. For determining aggregate size distribution, the sieve simulation and optimizing rules (SSOR) method was developed. Water to cement ratio was estimated by having volume ratios of cement paste components extracted from processing BSE images. Also, cement content was estimated using a proposed equation by having volume ratios of cement paste and water to cement ratio. Thus, using the proposed methods in this study, volume ratios of aggregates, cement paste, and air voids and aggregate size distribution, water to cement ratio, cement content, and water content can be estimated for an unknown hardened concrete. The statistical comparison between the actual mixture proportions and the results obtained from performing the proposed methods on samples made with three different aggregate size distributions and a range of cement content (300–500 kg/m3), and water to cement ratio (0.32–0.6) highlights the brilliant performance of the introduced methods.