Scalable ontology-based V&V process for heterogeneous systems and applications

Abstract

This work focuses on ongoing research within the EU-funded EnerMan project aiming at improving the energy efficiency of manufacturing systems. Industrial use cases are generally too constrained to easily proceed to the verification and validation (V&V) of the scientific approaches tackling their challenges. In this context, we propose an ontology-based framework with a methodology assessing the scalability of heterogeneous systems, environments, and missions in a V&V context. Indeed, projecting these industrial and laboratory applications onto a meaningful ontology allows them to be flattened out to the same scale from a semantic point of view. Reasoning is used to evaluate the extent to which a given scientific approach can be verified on a laboratory use case different from the industrial scenario on which it has to be validated. The framework has been implemented using Protégé and Owlready2, and applied to a scientific approach focused on a blind source separation technique used to identify system operating modes in a black box manner, tested on a coffee machine and two industrial case studies (a vehicle testbed's heating ventilation and air conditioning system, and a chocolate production line).