Enhancing Flood Routing Accuracy: A Fuzzified Approach to Nonlinear Variable-Parameter Muskingum Model

Abstract

Abstract Flood routing stands as a critical technique for water engineers in effectively managing and mitigating the consequences of floods. Among the prevalent hydrological methods, the Muskingum method emerges as a highly efficient approach, owing to its accuracy and simplicity in application. This research introduces a novel partitioning framework aimed at refining outcomes from a nonlinear variable-parameter Muskingum model. This improvement is achieved by introducing fuzzification to the boundaries of adjacent sub-periods. The results underscore the efficacy of the proposed method in enhancing the accuracy of routed outflow, aligning well with the inherent characteristics of a flooding event. Validation of the newly introduced fuzzified nonlinear variable-parameter Muskingum model was conducted using four distinct case studies from the literature. These encompassed Wilson's dataset, the flood events in Rivers Wye and Wyre, and Viessman and Lewis' data. The evaluation of the proposed framework's effectiveness utilized metrics such as the Sum of Squared Deviations (SSQ), the Sum of Absolute Deviations (SAD), Mean Absolute Relative Error (MARE), and the Variance Explained in Percentage (VarexQ). The results demonstrated a notable increase in the accuracy of the nonlinear Muskingum model for the respective cases studied. This implies that the proposed partitioning framework is adaptable to various flooding events, irrespective of their intensity and duration, thereby advancing the applicability of any variable-parameter Muskingum model.