Distributed sequential estimation algorithm for steel temperature in hot strip mill

This paper is concerned with the steel temperature estimation problem in hot strip mill. We focus on the steel bar temperature estimation during the rough rolling process with the aid of industrial wireless network for its high reliability communication and low power consumptions. However, the mobility of the steel bar and the complexity of the industrial environments have posed new challenges for the temperature monitoring. According to the steel bar's mobility, we propose a distributed sequential estimation algorithm to realize the continuous temperature tracking and gradually remove the estimation error. It consists of a cluster-based estimation algorithm and an initialized fusion algorithm, whose design are based on two criteria, namely, unbiasedness and optimality. Finally, simulation studies are presented to validate the performance of the proposed algorithm.