SENSE Technology Stack

The SENSE technology stack provides a modular, open framework for developing systems that generate meaningful, human-understandable explanations for events in cyber-physical systems. At the centre of the stack is the SENSE Semantic Model (SENSE Ontology), a machine-readable representation of domain knowledge and system structure that enables semantic interoperability and contextual understanding.

Sensor data from connected devices is continuously collected and enriched with contextual information from the ontology to support real-time monitoring of system behaviour. This enriched data is processed to detect high-level events—such as identifying that a window has been opened—and recorded within the Semantic Event-handling Module, which also incorporates causal knowledge captured in the ontology (e.g., recognising that an open window may lead to increased energy consumption). Building on this foundation, the Explainability Module interprets events and causal relations to generate clear, context-aware explanations. A conversational interface makes these insights accessible to end-users, allowing them to ask natural-language questions and interactively explore the explanations produced by the SENSE framework.

A key contribution of the SENSE project to the wider research and innovation community is the open release of the components that make up this technology stack. By making these resources publicly available, SENSE supports transparency, reproducibility, and collaboration across energy systems, smart buildings, and digital infrastructures. The source code is accessible through our GitHub repositories and includes:

  • SENSE Framework – the core software infrastructure that implements the architectural building blocks for integration, experimentation, and deployment.
  • SENSE Ontology – a structured knowledge model that enables data harmonisation and semantic interoperability across diverse systems.
  • Proof-of-Concept Implementations, demonstrating the practical application of the technology stack:
    • the Energy Community PoC, showing how SENSE supports energy management and community-driven energy sharing;
    • the Smart Building PoC, illustrating the role of SENSE in intelligent building operation and optimisation.

By publishing these open-source components, the SENSE project invites researchers, developers, and practitioners to explore, adapt, and extend the technology. This open approach accelerates innovation, fosters cross-disciplinary collaboration, and ensures that the project’s outcomes deliver long-term value beyond its immediate scope.

For further reading, please refer to the following paper:

[1] Ehrenmüller, K., Diwold, K., Schwarzinger, T., Steindl, G., Prüggler, W., Ekaputra, F. J., & Sabou, M. (2025, October). Enhancing Transparency in Smart Grids: the SENSE Framework. In International Semantic Web Conference (pp. 415-433). Cham: Springer Nature Switzerland.

SENSE Ontology

The SENSE Ontology, the concrete implementation of the SENSE Semantic Model, provides the semantic foundation that enables SENSE to deliver transparent, causal, and user-centred explanations in complex cyber-physical systems. It consolidates five core knowledge domains—topology, observation, causality, user context, and explanation—into a unified, machine-readable framework that supports semantic reasoning and interoperability across heterogeneous data sources.

Developed using the Linked Open Terms methodology and guided by real-world use cases in smart grids and smart buildings, the ontology addresses key gaps in traditional explainable AI approaches by modelling domain structure, system behaviour, and causal relations in a consistent and reusable way. Evaluation activities demonstrated strong performance in responding to stakeholder-driven queries, while pilot deployments confirmed its value in providing structured, actionable insights into CPS anomalies.

The SENSE Ontology is openly maintained and publicly available, inviting researchers and practitioners to adopt, reuse, and extend it within their own domains and applications.

For further reading, please refer to the following paper:

[2] Ehrenmüller, K., Ekaputra, F. J., Schwarzinger, T., Steindl , G., Shimizu, C., & Sabou, M. (2025). When Things Go Wrong: The SENSE Ontology for Explaining CPS Anomalies. Manuscript submitted for publication.

Further Details and Resources

For more in-depth information and reading, please refer to our project results website: https://sense-project.net/results/