Development and Evaluation of the Self-Adaptive Horse Algorithm (SA-HA) for Reservoir Operation Management

Articles in Press

Authors

Department of Civil Engineering, University of Qom, Qom, Iran

10.22034/ceej.2026.70140.2468

Abstract

Optimal water resource management under unstable climatic conditions requires advanced computational tools that can support intelligent, environment adaptive decision making. In this study, a self adaptive version of the Horse Algorithm (HA) is developed to alleviate the limitations of the classical version, such as premature convergence and sensitivity to parameter settings. In the proposed model, each agent (horse) is equipped with a dynamic learning rate that is automatically updated based on the improvement of the objective function at every iteration. This mechanism balances wide range exploration in the early stages with focused exploitation of promising regions in the later stages. To assess performance, the self adaptive Horse Algorithm (SA HA) and the classical HA were applied to the optimal water allocation problem of the Qaranqu Reservoir in East Azerbaijan Province, with the objective of maximizing the ratio of water supplied during the operation period. Results show that the adaptive version, compared with the baseline, yields an approximate 17 % increase in water supply efficiency, a 27 % reduction in deficit, and improves the system reliability index from 0.59 to 0.68. Moreover, SA HA, by dynamically and automatically tuning its movement parameters, reduces result volatility and accelerates convergence. The findings indicate that incorporating a self adaptive mechanism into HA markedly enhances the algorithm’s accuracy, stability, and flexibility for managing complex water resource systems, making it an effective approach for optimizing sustainable operation decisions.

Keywords

Main Subjects

Journal of Civil and Environmental Engineering

Articles in Press, Accepted Manuscript
Available Online from 31 January 2026

Files

Share

How to cite

Statistics