Hybrid optimization model for calibrating the HBV hydrological model

Abstract

In this study, a hybrid optimization model is proposed for calibration and verification of the semi-distributed Hydrologiska Byrans Vattenbalansavdelning (HBV) hydrological model. The proposed model consists of the mutual integration of the heuristic Ant Lion Optimization (ALO) and the Sequential Quadratic Programming (SQP) optimization approaches. In this integration, ALO performs the global exploration process to seek potential locations where global optimum exists and SQP performs a local search over these locations for precisely finding the optimum solution. This integrated model is then used to calibrate the parameters of the HBV model by maximizing the Nash-Sutcliffe efficiency (NSE) as the objective function. The applicability of the proposed hybrid optimization model is evaluated on Gediz River Basin, which is one of the most important river basins of Turkey. The identified results indicated that the proposed hybrid optimization approach provides quite successful calibration and verification results in terms of the calculated runoff values compared to the hybridized use of the Genetic Algorithms (GA) and Powell optimization approaches in HBV-Light software package. © 2023 IAHR – International Association for Hydro-Environment Engineering and Research.