Prediction and control of greenhouse temperature: Methods, applications, and future directions

Greenhouse cultivation is crucial for global food security and sustainable agricultural development. To maintain efficient greenhouse crop production, temperature management is the core. For a better understanding of the current status, methods, applications, and future directions for greenhouse prediction and control, this paper provides a comprehensive review of different technologies for greenhouse temperature management. The contributions cover three key aspects: (1) The role of different sensing techniques, the Internet of Things, wireless sensor networks, and multimodal data fusion technologies in supporting greenhouse temperature management; (2) Status and advantages of traditional models, artificial intelligent (AI)-based models, and hybrid models for greenhouse temperature prediction; (3) Applicable scenarios and limitations of different control methods, such as fuzzy logic, model predictive control (MPC), reinforcement learning (RL) and intelligent systems for greenhouse temperature control. Comparative analysis demonstrates that deep learning models excel in greenhouse temperature prediction, while fuzzy logic, MPC, and RL exhibit unique strengths in greenhouse temperature control. Despite the potential demonstrated by advanced AI technology in greenhouse temperature management, practical applications continue to encounter challenges such as model robustness, interpretability, and computational efficiency. To fully exploit the potential of AI, future research should focus on developing plant-centric AI models, exploring the application of AI in sustainable energy management of greenhouses, and developing digital twin models to promote the development of AI and greenhouse technology, creating a new paradigm for sustainable, resilient, and intelligent agriculture. This review offers a comprehensive perspective on optimizing greenhouse environment management, which is crucial to address challenges in food security and climate change.