순환여과양식시스템 내 유량변동에 따른 무지개송어(Oncorhynchus mykiss) 용존 산소 물질수지 모델 평가

다국어 초록 (Multilingual Abstract)

Digital Twin technology replicates the physical environment into a virtual space, enabling real-time sensor data integration for precise system monitoring, analysis, and predictive management. This study aimed to optimize flow rates and minimize energy consumption in freshwater Recirculating Aquaculture Systems (RAS) for rainbow trout (Oncorhynchus mykiss), as a foundational effort toward Digital Twin implementation. Initially, a thermal growth coefficient (TGC)-based model predicted daily fish biomass and feed requirements, assuming a fixed oxygen consumption rate of 0.25 kg O₂ per kg feed. However, significant discrepancies between predicted and actual oxygen consumption were identified experimentally. To enhance accuracy, a multivariate regression model was developed using daily feed supply (kg/day), average fish weight (g), and water temperature (℃) as independent variables, yielding the refined regression equation: y = -0.391 + 1.712·x₁ - 0.0096·x₂ + 0.0365·x₃ Validation over six weeks demonstrated that this improved model significantly enhanced prediction accuracy (R² = 0.673), considerably reducing prediction errors compared to the initial model. Additionally, dynamic flow adjustments using variable frequency drive (VFD) pumps reduced daily energy consumption from 57.6 kWh/day to 42.97 kWh/day, a 25.4% improvement. Energy consumption per kilogram of biomass gains also dropped notably from 45.47 kWh/kg to 33.92 kWh/kg, highlighting increased operational efficiency. This research confirms the potential of advanced predictive modeling within Digital Twin frameworks to optimize RAS operations, significantly enhancing energy efficiency and system sustainability.

목차 (Table of Contents)

• I. 서 론 1
• II. 재료 및 방법 5
• II-1. DO 소비 예측 및 제어 알고리즘 개발 5
• 가. 성장 예측 모델 도출 5
• 나. DO 물질 수지 모델 도출 7
• 다. DO 변동 예측을 통한 유량 제어 알고리즘 도출 11
• II-2. 사육 실증 15
• 가. 사육 시스템 15
• 나. 실험 어류 및 조건 20
• 다. 수질 모니터링 22
• II-3. 에너지 사용량 분석 23
• II-4. 통계 분석 25
• III. 결 과 26
• III-1. DO 농도 변화 27
• III-2. 어류 산소 소비량 30
• III-3. DO 물질 수지 모델 평가 32
• III-4. DO 물질수지 모델 개선 38
• III-5. DO 물질수지 모델 간 성능 평가 40
• III-6. 성장 및 수질환경 44
• III-7. 에너지 소비량 분석 48
• IV. 고 찰 50
• IV-1. DO 물질수지 기존 모델 50
• IV-2. DO 물질수지 개선 모델 53
• IV-3. DO 물질수지 모델의 향후 발전 방향 및 통합 고찰 56
• 요약 60
• 감사의 글 62
• 참고문헌 65