Development of a Two-Stage Method for Zoned Pore Pressure Clustering Using FCM and GMDH Models (Case Study: Eyvashan Earth Dam)

To identify the behavior and health monitoring of dams, it is necessary to correctly interpret the results of instrumentation in different phases of dam construction, impounding, and operation. Therefore, the approach based on spatiotemporal prediction is presented to improve the interpretation of pore pressure behavior of Eyvashan Dam. Using the results of other existing healthy piezometers, this research proposes a zoned pore pressure clustering two-stage distribution model using panel data that can reconstruct the missing data. In the first stage, the optimal spatiotemporal clustering of pore pressure change monitoring is obtained by designing a Fuzzy C-Means method (FCM), which will enable the monitoring of points of the dam where instrumentation has not been designed and installed. In the second step, the Group Method of Data Handling (GMDH) model has been used to predict pore pressure using points located in a cluster. The results of the statistical performance evaluation of the data show that the two-stage clustering model of FCM and GMDH algorithm, taking into account the time delay, can be a suitable and efficient method in modeling and predicting pore pressure. The proposed method will facilitate the detection of unusual areas of piezometric pressure and related safety diagnosis.