Automated Generation of a Digital Twin of a Process Plant by Using 3D Scan and Artificial Intelligence

Abstract

The simulation of production processes using a Digital Twin is a promising tool for predictive planning, analysis of existing systems or process-parallel monitoring. In the process industry, the concept of Digital Twin provides significant support for process optimization. The generation of the Digital Twin of an already existing plant is a major challenge – in particular for small and medium-sized enterprises. In this sense, the twinning of the existing physical environment has got a particular importance due to high effort. Shape segmentation from unstructured (e.g. point cloud data) is a core step of the digital twinning process for industrial facilities. This is an inherent issue of Product Lifecycle Management how to acquire data of existing goods. The practice of Digital Twin is described based on object recognition by using methods of Machine Learning. The exploration of the pipeline semantics presents a particular challenge. The highly automated procedure for the generation of Digital Twin is described based on a use case of a biogas plant. Commercial deployment, pitfalls, drawbacks and potential for further developments are further explored.

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

“…A fast, precise and robust object recognition in a brownfield environment is a prerequisite for digital twin and thus subject of continuous research. The affordability of object reconstruction with semantic enrichment must be justified primarily in the cost-benefit ratio compared to manual remastering [5]. If object recognition is successful, as is the case here in an application from the biochemical industry, the reasoning is obvious.…”

“…In a previous work, the approach of generation of the digital twin of a brownfield process plant based on automatic object recognition by machine learning was presented [5]. The next step in the evolution of digital twin provides the possibility of a continuous update of a digital twin under aforementioned changing operating conditions, wellharmonized with the updated P&ID [6].…”

“…The prerequisite for this is, first, an automatic object recognition system that is capable of recognizing objects such as pipelines or fittings on the basis of a fraction of the scanned points and determining their position in space [11]. This goal was achieved in the previous research [5] and the object classes can be recognized reliably by using methods of artificial intelligence (machine learning, ML) [12]. Secondly, it must be possible to enrich the recognized objects with the corresponding information in the P&ID with as little effort as possible [13].…”