The use of digital twins to overcome low-redundancy problems in process data reconciliation

In the digital era, data mining, analysis, and interpretation are of primary interest for industry 4.0. Processes provide a huge amount of dataset, which should be processed in the shortest possible time. Reliable process monitoring in real-time is of primary concern in the chemical industry due to its effect on the production costs as well as on decision making and performance, emissions, and safety monitoring. This can be achieved by exploiting the potential of simulations and digital twins to support data reconciliation (DR), gross error (GE) detection, and Dynamic Data Reconciliation (DDR). Data reconciliation is a numerical procedure used to correct measurements to fulfill material (i.e., mass flow rates and/or compositions) and energy balances, obtaining a coherent picture of the plant operating conditions. This chapter introduces a hardware and software platform for digital twin-aided data reconciliation and validation on a carbon capture industrial plant section.