Download PDFOpen PDF in browser

A combined model-based and data-driven approach for monitoring smart buildings

16 pagesPublished: January 6, 2018

Abstract

This paper combines a residual-based diagnosis approach and an unsupervised anomaly detection method to develop a hybrid methodology for monitoring smart buildings for which complete models are not available. The proposed method combines data mining approach and model-based diagnosis to update a diagnosis reference model and improve the overall diagnostics performance. To estimate the likelihood of each potential fault in complex systems like smart buildings, the dependencies between components and, there- fore, the monitors should be considered. In this work, a tree augmented naive Bayesian learning algorithm (TAN) is used for the classification. We demonstrate and validate the proposed approach using a data-set from an outdoor air unit (OAU) system in the Lentz public health center in Nashville.

Keyphrases: combined diagnoser, data-driven diagnosis, model-based diagnosis, residual analysis, Tree Augmented Bayesian classifiers

In: Marina Zanella, Ingo Pill and Alessandro Cimatti (editors). 28th International Workshop on Principles of Diagnosis (DX'17), vol 4, pages 21--36

Links:
BibTeX entry
@inproceedings{DX'17:combined_model_based_and_data_driven,
  author    = {Hamed Khorasgani and Gautam Biswas},
  title     = {A combined model-based and data-driven approach for monitoring smart buildings},
  booktitle = {28th International Workshop on Principles of Diagnosis (DX'17)},
  editor    = {Marina Zanella and Ingo Pill and Alessandro Cimatti},
  series    = {Kalpa Publications in Computing},
  volume    = {4},
  pages     = {21--36},
  year      = {2018},
  publisher = {EasyChair},
  bibsource = {EasyChair, http://www.easychair.org},
  issn      = {2515-1762},
  url       = {https://easychair.org/publications/paper/cWm2},
  doi       = {10.29007/g44l}}
Download PDFOpen PDF in browser