An Approach for the Incremental Update of a Digital Twin of a Process Plant

Abstract

Process plants have a long life cycle which often has started prior to the digital era. Most recently, the concept of digital twin provides significant support for process optimization and reducing the erroneous impact of human factors. The twinning of the existing plants or a portion of is necessary based on automated and accurate object acquisition by scanning. Automatic shape segmentation based on machine learning from unstructured (e.g. point cloud) data is a core step of the digital twinning process for industrial facilities. Once this twin has been generated, a frequent, mostly incremental update is necessary due to modifications during the maintenance and modernisation. The exploration of the pipeline semantics presents a further research step to keep coherence of 3D model with the pipeline and instrumentation diagram. In this paper, the entire approach is described based on a use case of a biogas plant in an industrial collaboration. Transdisciplinary aspects of this approach are further explored.