IGWO-AS: An Enhanced Grey Wolf Optimizer for Relay-Node Deployment in LoRa-Based Smart Greenhouse Systems

Greenhouse IoT systems often suffer from uneven sensing coverage and unreliable long-range communication in obstacle-constrained environments, which limits the stability of environmental monitoring and control. To address these issues, this study proposes an IoT-based closed-loop greenhouse monitoring and control system that integrates relay-node deployment optimization with system implementation. An improved Grey Wolf Optimizer, termed IGWO-AS, is developed by incorporating an adaptive phase adjustment mechanism (APAM), a stochastic differential disturbance exploration mechanism (SDDE), and a simulated annealing-inspired acceptance criterion (SAIAC), thereby enhancing global exploration, maintaining population diversity, and alleviating premature convergence in complex greenhouse scenarios. In addition, a three-layer architecture based on LoRa communication technology is constructed, consisting of the perception layer, network layer, and application layer, to support environmental sensing, reliable data transmission, and closed-loop regulation through an IoT cloud platform. Field experiments conducted in a 50 m × 50 m greenhouse with obstacles show that IGWO-AS achieves an average effective coverage rate of 93.5%, which is 6.3 percentage points higher than that of the standard GWO, while reducing the packet loss rate at 100 m to 0.2%. The deviation between simulated and measured coverage remains within 3.5%, indicating good consistency between simulation and practical deployment. In simulation studies across multiple scenarios, IGWO-AS further exhibits faster convergence and better solution stability than the compared algorithms. Overall, the proposed algorithm–system co-design framework demonstrates good practical feasibility for intelligent greenhouse monitoring and control.