Implementation of a UAV-sensory-system-based hazard source estimation in a chemical plant cluster

Abstract

Since there has been growing concern about the damage that atmospheric leakage and dispersion accidents may have done to human beings, researchers are dedicated to study effective and feasible source estimation methods in chemical plant clusters. In this paper, the safety monitoring of chemical production process is conducted via an unmanned aerial vehicle (UAV) monitoring system. Based on the observed data from this system, a source estimation method incorporating Bayesian inference and Particle Swarm Optimization (PSO) is proposed. Furthermore, the method is extended to discuss the source tracing algorithm using game theory in the UAV system. Finally, a practical case study is carried out to verify the feasibility and credibility of the proposed method. Results show that the method and system are helpful for safety monitoring and risk assessment in a chemical plant cluster.