Intelligent assessment method of progressive collapse resistance of corroded latticed shell structures considering time-varying effect

Abstract

Single-layer latticed shells feature high load-transfer efficiency and flexible modeling, making them a commonly used roof structure form in large public buildings such as gymnasiums and airport terminals. However, in recent years, accidents of progressive collapse of roof structures have occurred frequently, and the corrosion problem of structures during long-term service is one of the main causes of such accidents. Taking the K8-type spherical latticed shell as the research object, this paper considers the time-varying effect of corrosion, defines the latticed shell collapse criterion coefficient and collapse resistance degradation coefficient based on displacement response, and uses these coefficients to evaluate the progressive collapse resistance of the latticed shell and the degree of structural performance degradation under the influence of corrosion. The verified numerical simulation method is used to carry out the parametric analysis of the progressive collapse resistance of the corroded latticed shell. The results indicate that the progressive collapse resistance of latticed shells deteriorates with increasing service time. For a 40 m span shell, the resistance decreases by 70.2 % after 50 years compared to the uncorroded case, and progressive collapse occurs after 60 years. With the expansion of the corrosion area, the collapse resistance initially decreases and then tends to stabilize. The location of initial failure, shell geometry, and load distribution all strongly influence the deterioration of the latticed shell’s collapse resistance. On this basis, the data set of the deep feedforward neural network is generated, and the intelligent evaluation model of the progressive collapse resistance of the corroded latticed shell is established, which can accurately and efficiently predict the progressive collapse resistance of the corroded K8 latticed shells.