Abstract

Modern high-performance concrete is one of the most versatile, durable, and cost-effective building materials known to man. Its composition has been well-characterized in 'test sample' reports from laboratory specimens and trial castings. The compressive strength of concrete is the most common performance measure used by engineers when designing concrete structures (e.g. buildings, pipes, roads and bridges). However, concrete performance often reveals large differences from that of 'test sample concrete'. To resolve specific problems related to environmental hazards and constructional materials a new approach, including methods for near-real-time analysis, is required. In this study, diffuse reflectance spectroscopy was used across the visible, near- and shortwave-infrared spectral regions (400 to 2500 nm) as a tool to assess the strength of high-performance concrete in situ. The suggested spectral model was constructed as a data-mining method that enables differentiating sample reflectance and extracting quantitative information. To examine the potential of this model for predicting compressive strength, several controlled experiments were conducted in which concrete samples were spectrally measured and simultaneously tested for compressive strength. Spectral analysis provided accurate predictions of concrete strength. Since low-cost, rapid methods are required; this might be an ideal tool for concrete-strength estimation in situ in near real-time, and warrants further study.

Key words: High-performance concrete, compressive strength, diffuse reflectance spectroscopy, spectral model, non-linear iterative partial least squares (NIPALS) analysis.