Digital twins are virtual counterparts of physical systems, crucial for monitoring and reasoning in cyber-physical systems. Implementing a Digital Twin requires a language specification for ensuring interpretability and interoperability among integrated systems. Ontologies from various domains can address syntactic and semantic interoperability issues. To support global communication and knowledge exchange, mechanisms for system integration are needed. This paper proposes a model-based approach for automatically generating a modular multimodal core ontology specific to a port Digital Twin. The approach involves establishing syntactic and semantic alignments between ontologies using various algorithms, followed by a refactoring step to unify concepts for the domain. Finally, reasoning services are applied to verify the consistency of the generated ontologies.