Crack prediction based on curvature-crack spacing model and probabilistic theory

Abstract

Crack prediction of reinforcement concrete infrastructures in long-term service is critical for guiding daily maintenance and strengthening of structures and providing timely alarm for possible accident. This paper proposes a probability-based crack prediction method to presage bending-induced cracking-prone regions and the corresponding cracking probabilities. Firstly, the critical equation, which is used to describe the relationship between curvature and crack spacing, is reduced based on bond-slip theory of reinforcement and concrete, and the critical equation is divided into four cases according to the relative position of the shear stress peak when a new crack occurs. Secondly, a probability-based crack prediction method is proposed based on the probability and statistics principles. The viability of the proposed method is examined using two reinforced concrete (RC) beam specimens subjected to four-point bending. The data processing results indicate that the method successfully divides several cracking-prone regions based on the curvature obtained from the initial stage of loading, and the prediction result is highly consistent with the actual crack positions, which demonstrates the great potential of the proposed crack predicting method.