Goal of the Digital Twin process

The digital twin process has two main goals:

1. Early virtual commissioning (SoftFAT)

   • Testing the control system and communication without real hardware.

   • Verification of sequences, signal flows and parameter behavior.

2. Long-term learning and optimization process

   • After commissioning, the digital twin is supplemented with real operational data.

   • This creates a learning, data-based model of the process behavior.

Target vision:

The digital twin according to Selmo is initially a deterministic emulation of the hardware and evolves with data and experience into a mathematically founded, behavior-capable process model.