Development and Experimental Validation of a Cyber-Physical System for Adaptive Control in Pond Aquaculture

Abstract

This paper presents the development and experimental validation of a cyber-physical control system designed for pond aquaculture operating in open and dynamically changing environmental conditions. The study aims to enhance aquaculture control efficiency through automated monitoring and adaptive control of key water quality parameters. An analysis of the primary causes of high fish mortality in pond farms is conducted, highlighting the limitations of traditional manual monitoring approaches. A model of a cyber-physical system is proposed, integrating physical, cyber, and control subsystems with the external environment, accompanied by a formal mathematical description of their interactions. Key controlled parameters – water temperature, pH, and dissolved oxygen concentration –are defined, and a correlation analysis is performed on monitoring data. An intelligent monitoring data analysis method, based on multi-criteria decision-making and adaptive control, is implemented using IoT sensor data. A working prototype of the system was deployed and tested at operational pond farms in the Volgograd region. Experimental results demonstrate a significant reduction in monitoring time (up to 95%), an 86% decrease in the frequency of abnormal conditions, and a 9% reduction in feed consumption compared to conventional management methods. The findings confirm the effectiveness, reliability, and practical applicability of the proposed cyber-physical approach for improving the sustainability and productivity of pond aquaculture systems.