Study the effect of 2D particle shape and size on angle of repose using a new detection algorithm

• A new angle of repose detection algorithm using image processing techniques. • Four different 2D shape indices are calculated and compared. • Dry sand angle of repose is related linearly to particle shape and size. • Internal friction angle can be predicted using the angle of repose. The angle of repose (AoR) is a critical parameter characterising the flowability and friction properties of granular materials. The objective of this paper is to develop an image-based method for calculating the angle of repose and to analyse the relationship between particle micro-shape and the macro accumulation state. Six types of sand with different particle sizes were tested to determine their AoR values using the well-established fixed funnel method, and a fitting algorithm for sand pile boundary is proposed in this study. Further, a Lasso-based multiple regression model was developed that incorporates roundness and particle width as independent variables. The AoR results indicated that, among the various 2D shape indices, roundness exhibited the strongest correlation with AoR. In addition, the AoR values exhibited a monotonically decreasing trend with decreasing particle size. The AoR values were compared to the internal friction angles obtained from the same sand samples, and the correlation coefficient between the internal friction angle and the AoR was 0.898. This study demonstrates that there is a strong correlation between the internal friction angle and AoR of the dry sand tested in this study.