Authors:

Kayla Xu

Date Created:

Not specified

Course Title:

Professor:

Not specified

About Paper:

Increasing volumes of energy related sensing data are created by diverse sources such as IoT enabled smart manufacturing facilities and environment monitoring sensors. It enables researchers to develop new data- driven machine learning models to study building energy consumption, prediction, and optimization. However, researchers face multiple challenges in querying, accessing, and processing these sensing data. For example, building/energy sensing data typically come with diverse, non-standardized metadata, making it hard for users to correctly understand the semantic attributes, relationships, and structures of the data. As a first step to tackle this metadata challenge, my project aims to develop a novel online metadata visualization and exploration tool that enables researchers to intuitively discover, interact, and investigate the organization, relationship, and hierarchy of large amounts of sensing data for building energy efficiency. It is funded by the AnalytiXIN initiative sponsored by the Central Indiana Corporate Partnership. This tool was developed in Python using Jupyter Notebook. It retrieves the metadata of diverse building/energy sensing data from real-world manufacturing buildings from AnalytiXIN's Data Lake via a data query API. The metadata is then transformed into a graph model with nodes and edges. The ipcytoscape library is leveraged to create a dynamic and interactive graph representation of the hierarchical structures and relations among the data. Users can view and interact with the graph representation intuitively. This tool is generic and can be easily adapted to work with sensing data from other domains beyond the manufacturing energy community, making data more findable, accessible, interoperable, and reusable (FAIR).

Source:

Purdue University / 2023

Topics:

No topics listed

Co-authors:

Kayla Xu