Optimization of the performance of riprap apron to protect spur dikes against scouring using Taguchi and grey relational analysis methods

One of the most significant problems faced by hydraulic engineers is the local scouring of in-stream structures. The optimal design of scour protection methods requires time consuming, laborious and costly studies. Fractional factorial methods have been provided to formulate experimental designs which can optimize designs with fewer experiments and high accuracy. Therefore, this study aimed at using Taguchi and grey relational analysis (GRA) methods to investigate the effect of three factors: (1) spur dike angle, (2) size of riprap stones, and (3) riprap extent on the local scouring. An additional goal was to achieve an economical pattern for the riprap apron so that it delivers the best performance in protecting spur dikes with different angles. The results show that the riprap apron decreases the scouring depth and moves the scouring hole away from around the structure. However, increasing the spur dike angle decreases its performance in scouring reduction. Increasing the area of the riprap apron protected the downstream walls better and reduced the edge failure of the riprap; however, the erosion in the opposite wall increased. In addition, analysis of variance showed that the riprap extent and the spur dike angle, respectively, have the highest effects on the riprap performance, respectively. The comparison of the predictions of the Taguchi and GRA methods with observed values showed that these two methods have high capabilities to predict the results of scouring studies.