Estimation of Bridge Pier Scour using Adaptive Neuro-Fuzzy Inference System Optimized with Imperialist competitive algorithm

Determining the scouring depth around bridge piers during flood has been considered as one of the main and important parameters in designing the foundations of bridge piers. Thus, it has been the subject of numerous studies to provide precise methods to calculate these elements. Due to the high complexity of this phenomenon and various involved parameters, the utilized empirical relations have not sufficient accuracy and efficiency; accordingly, economic and technical design based on their results is not feasible. In this regard, a new model is provided to estimate the scouring depth around bridge piers using Neuro-Fuzzy Comparative Inference System that optimized by Imperialist Competitive Algorithm. Comparing the results of proposed model with the results of base models showed that the accuracy and efficiency of the base models increased significantly with these optimizations. In addition, comparing the results of proposed models with the results of empirical relations from Mississippi, HEC-18, Laursen & Toch, and Froehlich showed that proposed model has significantly better accuracy in comparison with mentioned experimental relations.