Are you an EPFL student looking for a semester project?
Work with us on data science and visualisation projects, and deploy your project as an app on top of Graph Search.
Industrial plant data are difficult to find in academic literature for a number of reasons such as confidentiality, and thus intentional masking, or problem size reduction. These common practices limit the ability of researchers to apply novel methods to real cases and understand energy consumption of real industrial plant instances. This is especially pertinent in the field of process integration, as realistic representations of real processes form the basis for the application of novel technologies. Few efforts have been made in this area, demonstrating the added value of these profiles; thus, a clear methodology is required for constructing such energy consumption profiles. The method proposed in this work defines an approach for constructing the heat profiles of major industries in a generic way. Parameterized models of several major European industries are presented for defining specific production/plant instances based on contextual specificities to represent different production pathways. The profile construction methodology is described for several situations of data access. Confidentiality issues are addressed by different anonymization techniques such as aggregation, statistical treatment, or by using data which are already publicly available. In this work, data were gathered from real plant operations and validated at higher levels using public information. Although the potential applications and implications of these profiles are clear, two cases are presented to exhibit adaptation of the parameterized models to specific instances and profile use for process integration problems. Varying the model parameters represents different plant instances and thus yield different integration solutions for the major process industries included in this work.
Brice Tanguy Alphonse Lecampion, Andreas Möri
Brice Tanguy Alphonse Lecampion, Andreas Möri
Andrew James Sonta, Yufei Zhang