Swarm intelligence‐based optimal device deployment in heterogeneous Internet of Things networks for wind farm application

Summary Internet of Things (IoT) is becoming a suitable surveillance system in modern power grid, right from power generation to distribution. In the case of wind power generation application, the entire wind farm is considered completely monitored using edge devices known as IoT devices. Device deployment is a major obstacle that can be addressed for this application as unfair device deployment can lead to connectivity issue. This study has two main objectives. The first objective is to determine the optimal number of IoT devices that need to be place in the potential key elements of the wind turbines/targets for coverage assurance. The second objective is to determine the optimal number of relay devices. Spacing wind turbines several hundred meters apart brings the connectivity as a second critical issue, which can be handling by placing optimal number of relay devices. Hence, the focus of this study is to address connected target coverage problem in optimal device placement of heterogeneous IoT network for wind farm monitoring. This paper discusses in detail the swarm intelligence technique known as particle swarm optimization (PSO) to solve the connectivity issue and to minimize the cost of devices deployment. Additionally, a hybrid swarm intelligence model called hybrid PSO‐ACO (HPACO) algorithm is proposed. The main goal of this hybrid model is to reduce the number of relay devices, minimize deployment cost, and to improve network connectivity. The performance of both PSO and HPACO algorithms are validating by comparing the extensive simulation results with ant colony optimizations (ACO) algorithm and other approximation algorithm. Coverage assurance: Optimal IoT device placement in the potential key elements of the wind turbines/targets for coverage assurance. Connectivity assurance: Connecting all the targets with the base station through relay node placement. Optimal relay node placement using swarm intelligence technique such as ACO, PSO, and hybrid PSO‐ACO (HPACO) algorithms.