Agro ilm Том 124 № 3 (2026) · с. 74-76
INTELLIGENT STOCHASTIC CONTROL OF CLOSED WATER SYSTEMS USING RELAXATION-BASED HYDRODYNAMIC MODELS
Ismailov, Mirhalil, Ismailov, Otabek, Ozodov, Ezozbek
Аннотация
This paper addresses the challenge of automated control for closed water systems (e.g., endorheic lakes) subject to stochastic hydrological dynamics and internal relaxation processes (non-Markovian behavior). We propose an intelligent stochastic control framework based on Relaxation-Stochastic Models (RSMs) that capture both probabilistic external drivers and system memory. A risk-constrained optimal control problem is solved in real-time using Receding Horizon Control integrated with an LSTM metamodel of the RSM. Applied to the Aydar Lake system in Uzbekistan, the framework reduces the probability of breaching critical ecological levels by 37% over a 10-year horizon while improving irrigation supply reliability by 18% compared to conventional strategies. The results demonstrate a robust, adaptive solution for sustainable water resource automation under deep uncertainty.
Stochastic control, closed water systems, relaxation processes, intelligent automation, LSTM, Aydar Lakeстохастическое управление, замкнутые водоёмные системы, релаксационные процессы, интеллектуальная автоматизация, LSTM, озеро Айдаркульstoxastik boshqaruv, yopiq suv tizimlari, relaksatsion jarayonlar, intellektual avtomatlashtirish, LSTM, Aydar ko‘li
Источник метаданных: OAI-PMH архив журнала · Sindex не хранит полный текст, а даёт ссылку на источник.