Application of ANFIS and LR in prediction of scour depth in bridges

This study uses the Adaptive Network-based Fuzzy Inference System (ANFIS) as a modeling tool to predict the scour depth in bridges. Experiments involved different sediment sizes, flow rates, and time evolution with and without Epipremnum aureum for predicting the effects of scouring on integral bridge piers. A total of 2500 data were taken at the maximum location of scour, and 17,500 scour data were taken at every pile for each time interval. Single row and double row pile integral bridge piers with pile group model were embedded in the two floodplains. The input data and its corresponding scour depth in bridges as output data were recorded at all testing stages. Results from ANFIS were compared with the classical linear regression (LR). ANFIS's results were highly accurate, precise, and satisfactory.