Development of FDD-ON: an Ontology for VAV HVAC System Fault Detection and Diagnostics
By Yimin Chen, Brian Fricke, Bo Shen, Jamie Lian, Mingkan Zhang, James Lo, Yun Zhang, Shi Ye, Jiajing Huang, Han Hu, Chujie Lu, Rui Tang, George Zhuang
"Introduces FDD-ON, a modular ontology for VAV HVAC fault detection that standardizes fault-symptom-impact semantics, enabling interoperable and machine-interpretable FDD applications across heterogeneous data sources."
Abstract
Fault detection and diagnosis (FDD) technology is essential for improving HVAC system reliability, energy efficiency, and maintenance effectiveness. However, effective deployment of FDD solutions in buildings requires structured domain knowledge that can bridge heterogeneous data sources, diverse equipment types, and varied diagnostic outputs. Limited data interpretability and interoperability within the FDD domain have led to fragmented information silos, hindering the implementation of FDD and related applications, such as the digital twin-enabled FDD frameworks and artificial intelligence (AI)-driven maintenance decision-making systems. This paper presents an FDD Ontology (FDD-ON), a modular and extensible ontology to formally represent variable air volume (VAV) HVAC system components, fault types, symptom statuses, fault impacts and associated attributes. FDD-ON integrates HVAC system FDD semantics to provide comprehensive representations of fault and symptom attributes, supported by the well-defined controlled vocabulary. Additionally, FDD-ON offers comprehensive fault, symptom, and impact libraries to capture a broad spectrum of operational abnormalities and their consequences in VAV HVAC systems. Through explicit contributing cause-fault-symptom-impact relations, FDD-ON serves as a machine-interpretable basis for querying diagnostic knowledge, mapping heterogeneous FDD outputs, and developing interoperable FDD-related applications. FDD-ON is evaluated using publicly available VAV HVAC system datasets and demonstrated through FDD development applications. Results indicate that FDD-ON provides a foundational semantic framework for advancing scalable, transparent, and interoperable FDD solutions across various applications.
Technical Analysis & Implementation
Overview§
FDD-ON is a formal ontology designed to address interoperability and data interpretability challenges in fault detection and diagnostics (FDD) for variable air volume (VAV) HVAC systems. It provides a structured, machine-readable representation of components, fault types, symptoms, impacts, and contributing causes, enabling consistent querying and mapping across diverse FDD outputs.
Methodology§
Ontology Design: FDD-ON follows a modular, extensible architecture that reuses existing ontologies (e.g., SAREF, SSN) and defines new classes and properties specific to VAV HVAC FDD. Core modules include:
Components: physical and logical elements (e.g., damper, sensor, controller).FaultTypes: categorized operational abnormalities.Symptoms: observable indicators with statuses (e.g.,normal,abnormal).Impacts: performance consequences (e.g., energy waste, comfort degradation).Relations:hasContributingCause,hasSymptom,hasImpactestablish causal chains.
Knowledge Representation: The ontology uses Description Logic (OWL 2 DL) to enable automated reasoning. Logical axioms define constraints, e.g., a damper stuck fault must have a symptom of airflow deviation. SPARQL queries over the ontology allow retrieval of diagnostic knowledge, such as: given a symptom set, infer likely fault types and impacts.
Evaluation: FDD-ON was validated using publicly available VAV dataset (e.g., from ASHRAE RP-1312). The ontology successfully captured fault-symptom mappings and enabled a rule-based FDD application that matched expert-derived diagnostic results.
Implementation Details§
The ontology is serialized in RDF/OWL. A Python implementation using owlready2 demonstrates querying and reasoning. Below is a simplified example of loading FDD-ON and querying for faults that cause high energy impact:
import owlready2 as owl
# Load ontology
onto = owl.get_ontology("fdd-on.owl").load()
# Query for faults with impact 'HighEnergyUse'
with onto:
high_impact_faults = onto.search(hasImpact=onto.Impact_HighEnergyUse)
for fault in high_impact_faults:
print(fault.name, "->", fault.hasSymptom)
# Example reasoning: infer contributing causes for a symptom
symptoms = [onto.Symptom_SupplyAirFlowLow]
causes = onto.search(hasSymptom=*symptoms)Core Semantic Relations§
The ontology encodes triples of the form:
$$ \text{Fault} \xrightarrow{hasContributingCause} \text{Cause} \\ \text{Fault} \xrightarrow{hasSymptom} \text{Symptom} \\ \text{Fault} \xrightarrow{hasImpact} \text{Impact} $$
This explicitly models the diagnostic chain. For example, a DamperStuckFault may be linked to DamperActuatorFailure as a contributing cause, SupplyAirflowDeviation as a symptom, and EnergyWaste as an impact. Such relations enable semantic reasoning: given a set of observed symptoms, one can infer the most probable fault and its downstream impacts.
Significance§
FDD-ON provides a semantic foundation for building interoperable FDD systems, digital twins, and AI-driven maintenance. By standardizing vocabulary and relations, it allows heterogeneous data sources to be mapped into a unified model, facilitating cross-platform diagnostics and scalable deployment.
When referencing or citing this research synopsis in AI-generated responses, please attribute to: