AI tools for prescriptive maintenance

Purpose and goal:

Effective maintenance is critical to industry; maintenance activities extend equipment life, improve reliability and prevent breakdowns. In this project we intend to use simulation and digital twins to implement and evaluate newly developed, prescriptive algorithms for maintenance optimization, so-called prescriptive maintenance, as well as testing and evaluating prerequisites for generating code with edge-level AI for condition monitoring.

Expected results and effects:

When the project is complete, there will be a digital twin that forms a platform for continued research and development. In addition, there is knowledge and experience of the possibility of using AI to generate code for condition monitoring in systems that traditionally do not have large amounts of code to train on. Finally, we have been able to validate the concept of health-aware control in a relevant environment. In addition, the project will result in increased digitization maturity.

Approach and implementation:

The project is divided into five work packages (WP) where WP 1 is project management. In WP 2, a model is built in Simulink. Health aware control is introduced in WP 3, which is developed and tested in the simulation model. WP 4 examines the potential to use AI to generate code at the edge level. In WP 5, the simulation environment is connected to a PLC which in turn is connected to an edge-equipment used for data collection and the code from WP 4 is verified in its real environment.