Information transfer from electrical design to simulation models in Modelica for virtual commissioning
Editor: Assoc. Prof. Poul Kyvsgaard Hansen, Professor John Rasmussen, Assoc. Prof. Kaj A. Jřrgensen, Assoc. Prof. Christian Tollestrup
Author: Zafrov, Radoslav; Eigner, Martin; Baudisch, Thomas
Institution: 1: Aalborg University, Denmark; 2: Design Society, United Kingdom
With the industry focusing on adaptive production, the complexity of production processes and production systems is constantly growing. Therefore, efficient interdisciplinary modeling and validation within the engineering process of manufacturing plants is now more relevant than ever. One computer assisted validation method of increasing importance for plant engineering is the virtual commissioning. Virtual commissioning is a method to commission and functionally test the control code running on the real control hardware against a model of the plant. This validation approach can be time-saving, since commissioning and test of the controller software does not have to be performed on the spot, after the plant has been physically built and set up. On the other hand the setup of the virtual commissioning model is a manual process today that involves a lot of time. The model setup requires results from the preceding mechanical and electrical design as input. The generation of these simulation models is done in designated simulation tools. Following the model based design paradigm, an approach has been developed, on how to automatically generate parts of multidisciplinary plant simulation models by re-using information from electrical design models. The goal of the approach is to avoid redundant manual re-modeling of aspects of the plant that have already been defined in electrical design. The considered simulation models are discrete behavioral models created using the Modelica language. A prototype application is presented in the paper, which has been programmed to validate the approach of translating electrical design models to a sub model of the virtual commissioning overall model using real plant data. A discussion on the applicability of this approach in real engineering scenarios concludes the paper.