Probing of the buckling of spherical shells with a distribution of imperfections

The buckling capacity of shell structures is strongly sensitive to imperfections and accurately predicting the critical buckling condition requires a priori knowledge of the defects. A novel procedure has been recently proposed to predict the buckling capacity of thin cylindrical shells based on the poking of pre-compressed shells to estimate their resistance to buckling. Subsequent studies revealed that the poking location plays a crucial role; determining the buckling capacity is only possible if the probing is done near the location of the most significant defect. However, the defects are generally unknown and often difficult to identify. We investigate the buckling of spherical shells containing a random distribution of defects. Using finite element simulations, we perform a statistical analysis of these imperfect shells, poking them with various indenters and at random-chosen locations. Such shells are more realistic and practically relevant than the previous single-defect cases. We believe our results on the probabilistic characterization of the stability of randomly imperfect shells using a statistical poking technique can inform the design rules of thin-walled structures.