Intelligent monitoring and control of farmland based on edge-cloud collaboration and digital twin for digital energy management: investment benefit analysis

• Published in: Renewables : wind, water, and solar Vol. 12; no. 1; pp. 43 - 13
• Main Author: Liu, Zheng
• Format: Journal Article
• Language: English
• Published: Heidelberg Springer Nature B.V 01.12.2025; SpringerOpen
• Subjects: Accuracy; Adaptive algorithms; Agricultural equipment; Agricultural land; Agricultural resources; Bottlenecks; Collaboration; Crop diseases; Crop yield; Data analysis; Data collection; Decision making; Deep learning; Deep reinforcement learning; Digital energy management; Digital twins; Economic benefits; Edge computing; Energy consumption; Energy flow; Energy management; Energy storage; Energy utilization; Environmental monitoring; Game theory; Intelligent monitoring of farmland; Investment benefit analysis; Irrigation; Irrigation water; Latency; Maintenance costs; Monitoring systems; Network latency; Optimization; Photovoltaic cells; Photovoltaics; Reinforcement; Resource allocation; Sensors; Systems stability
• DOI: 10.1186/s40807-025-00179-7
• ISSN: 2731-9237

The current farmland energy management and monitoring system still has problems, such as poor real-time data collection, low energy utilization efficiency, and insufficient intelligent decision-making. Focusing on digital energy management, this paper proposes a data collection and analysis based on edge computing and cloud collaboration architecture to improve the accuracy and real-time performance of farmland environmental monitoring. In terms of intelligent control, deep reinforcement learning is used to optimize irrigation decision-making, and adaptive algorithms are combined to improve the flexibility of agricultural equipment scheduling. Regarding energy management, a digital twin model of the photovoltaic energy storage system is constructed to achieve accurate prediction and optimization of energy flow. Edge-cloud collaborative architecture for real-time data collection/analysis, reducing network latency by 40% compared to traditional cloud-only models; deep reinforcement learning (DRL)-driven irrigation optimization, achieving 51% crop yield increase and 18% water efficiency improvement; digital twin modeling of photovoltaic-energy storage systems, enhancing energy flow prediction accuracy to 98.2% and reducing energy waste by 9.5%; game theory-based resource allocation to balance energy supply–demand, improving system economic benefits by 15%. The system stability reached 96.24%, and the maintenance cost was reduced by 21.0%. The utilization rate of irrigation water increased from 76.9% to 43.0% by 1.8 times, reaching 77.4%.