Codesign of Industrial Wireless Sensor Networks and Consensus-Based Sequential Estimation for Process Industries

Industrial wireless sensor networks (IWSNs) have been considered as promising technology to enhance target tracking and state monitoring in process industries with harsh environments. In this article, the codesign of IWSNs and consensus-based sequential estimation (CSE) for typical long-belt process industries is proposed. Specifically, a group-based IWSNs deployment strategy is first designed to cover the transportation belt. Over the group-based IWSNs strategy, the CSE algorithm is then proposed to conduct target tracking and state estimation using the distributed Kalman filter. To reveal the interrelation of IWSNs and CSE, the upper bound on the error violation probability of state estimation is first deduced from the sequential estimation process. Then, the algorithm is developed for the determination of IWSNs parameters (such as the number of groups of the IWSNs and communication design) and the CSE algorithm parameters (such as iterations of sequential estimation and estimation accuracy prediction). A case study of slab temperature monitoring over the hot strip milling process demonstrates the effectiveness of the proposed codesign of IWSNs and CSE.