Digitizing existing plants more efficiently with artificial intelligence

The documentation of the state of a plant is not only crucial when it comes to modernization measures. It also enables the use of 3D models for applications such as virtual operation or training purposes. Identified plant components also facilitate the use of augmented reality to support maintenance activities. For these purposes, the 3D models must correspond to the as-is-status, which in practice is often only the case with new plants.

Plant engineering companies today have to go to considerable lengths to digitize existing plants. A manually remodeling on the basis of 3D scan data is necessary. The piping systems of large plants alone can consist of several thousand pipe sections. Accordingly, digitization becomes time-consuming and cost-intensive. A particular challenge in this context is the correct identification and assignment of the piping and other equipment, such as pumps. The logic of the plant structure and the meta-information are stored in the corresponding P&ID-diagrams and must be linked to the 3D models.