Overview
The Ontology Alignment Project (OAP) is an open-sourced ontology used for modeling disparate sources of data and is the underlying framework for our Integrated Data Layer - Connect AI. It serves as the backbone of all of Noda’s data models and plays a key role in making data from disparate sources interoperable, understandable, and usable in a unified framework.
Core Objectives
Standardization: Ensure consistent definitions and relationships of concepts (e.g., "Temperature", "Device", "Zone") across platforms.
Interoperability: Enable seamless data exchange and integration between systems.
Scalability: Support rapid onboarding of new datasets or clients by reducing the need for bespoke mapping.
Open: Allow anyone to view and interact with the ontology via the website.
Alignment: Harmonize with other existing building ontologies such as Brick, Haystack, and DBO (Digital Buildings Ontology) to facilitate data portability between platforms.
Key Components
Canonical Taxonomy: A master vocabulary and structure. The fundamental types of “things” are sites, equips, points, attributes, KPIs, and units. Each type is further subdivided into a hierarchy.
Ontology: Definitions for how entities relate to each other, covering relationships like location, flow, containment, and metering.
Mappings: Defined linkages between terms in external ontologies (e.g., Project Haystack or Brick) and the canonical taxonomy & ontology.
Validation Tools: Ensure that new entities added conform to standards and comply with established rules.
Definitions
OAP: Ontology & taxonomy — sets the rules and the standards.
Data Model: An instance of a labeling system for the data.
Integrated Data Layer: A pipeline for the data to travel through. The schema for the Integrated Data Layer is built on the same concepts as the OAP and leverages the data model to serve up the data.
Importantly, the OAP is not just a “naming” standard, but rather a wrapper around the data that describes what any given entity is.
Use Cases
Data Ingestion Pipelines: Normalize data as it's ingested from various sources, allowing for unified analytics.
KPI Engines and Dashboards: Drive consistency in metric calculation by ensuring all entities refer to the same conceptual definitions.
AI Readiness: Provide semantically clean input data to machine learning models.
Visualization: Display data effectively in charts and widgets.
Rules & Insights: Indicate how equipment or a building is performing.
Automated Optimized Controls: Enable intelligent automation for system and energy performance.