Real-time Validation of Digital Models for Manufacturing Systems: a Novel Signal-processing-based Approach

Abstract

Recently, the connection between manufacturing systems and their digital counterparts has become of great significance for planning and control activities in a short-term scope. However, the alignment of a digital model with a very dynamic system is not always guaranteed, and traditional validation techniques cannot be used since they are designed for off-line simulators and rely on the availability of a large amount of data. This work develops a novel validation procedure inspired by signal-processing theory and a novel approach called quasi Trace Driven Simulation. The procedure is coherent with a Real-Time Simulation framework since it does not require large datasets to provide a good solution. The approach has been tried on test cases which demonstrated its applicability to a manufacturing environment.