Design and development of IIoT-based system for behavior profiling of nonlinear dynamic production systems based on energy flow theory

Abstract

In this paper, a solution to effective energy consumption monitoring of fast-response energy systems in industrial environments was proposed, designed, and developed. Moreover, in this research, production systems are characterized as nonlinear dynamic systems, with the hypothesis that the identification and introduction of nonlinear members (variables) can have a significant impact on improving system performance by providing clear insight and realistic representation of system behavior due to a series of nonlinear activities that stimulate the system state changes, which can be spotted through the manner and intensity of energy use in the observed system. The research is oriented towards achieving favorable conditions to deploy dynamic energy management systems by means of the Internet of Things and Big Data, as highly prominent concepts of Industry 4.0 technologies into scientifically-driven industrial practice. The motivation behind this is driven by the transition that this highly digital modern age brought upon us, in which energy management systems could be treated as a continual, dynamic process instead of remaining characterized as static with periodical system audits. In addition, a segmented system architecture of the proposed solution was described in detail, while initial experimental results justified the given hypothesis. The generated results indicated that the process of energy consumption quantification, not only ensures reliable, accurate, and real-time information but opens the door towards system behavior profiling, predictive maintenance, event forensics, data-driven prognostics, etc. Lastly, the points of future investigations were indicated as well.