Performing Diagnosis in Building with Partially Valid Heterogeneous Tests

Authors: Houda Najeh, Mahendra Pratap Singh, Stéphane Ploix, Antoine Caucheteux, Karim Chabir, Mohamed Naceur Abdelkrim

Abstract:

Building system is highly vulnerable to different kinds of faults and human misbehaviors. Energy efficiency and user comfort are directly targeted due to abnormalities in building operation. The available fault diagnosis tools and methodologies particularly rely on rules or pure model-based approaches. It is assumed that model or rule-based test could be applied to any situation without taking into account actual testing contexts. Contextual tests with validity domain could reduce a lot of the design of detection tests. The main objective of this paper is to consider fault validity when validate the test model considering the non-modeled events such as occupancy, weather conditions, door and window openings and the integration of the knowledge of the expert on the state of the system. The concept of heterogeneous tests is combined with test validity to generate fault diagnoses. A combination of rules, range and model-based tests known as heterogeneous tests are proposed to reduce the modeling complexity. Calculation of logical diagnoses coming from artificial intelligence provides a global explanation consistent with the test result. An application example shows the efficiency of the proposed technique: an office setting at Grenoble Institute of Technology.

Keywords: Heterogeneous tests, validity, building system, sensor grids, sensor fault, diagnosis, fault detection and isolation.

Digital Object Identifier (DOI): doi.org/10.5281/zenodo.3566319

References:

[1] Crawley, F. and Tyler, B . (2015). HAZOP: Guide to Best Practice.
[2] Galar,D,; Thaduri, A,; Catelani, M,; and Ciani, L. Context awareness for maintenance decision making: A diagnosis and prognosis approach. Measurement 2015, vol. 67 0263-2241.
[3] L. Scanu, S. Ploix, P. Bernaut and E. Wurtz (2017). Towards new model based energy management services. In IBPSA San-Francisco.
[4] Najeh, H., Singh, M. P., Chabir, K., Ploix, S., & Abdelkrim, M. N. (2018). Diagnosis of sensor grids in a building context: Application to an office setting. Journal of Building Engineering, 17, 75-83.
[5] Ploix, S. Des systèmes automatisés aux systèmes coopérants: application au diagnostic et à la gestion énergétique. Habilitation à diriger des recherches (HDR) 2009, chapter 3 and 5, 43-51.
[6] Reiter, R. A Theory of Diagnosis from First Principles. Artificial Intelligence 1987, vol. 32 (1) 57-95.
[7] Singh, Mahendra, Stéphane Ploix, and Frédéric Wurtz (2016). Handling Discrepancies in Building Reactive Management Using HAZOP and Diagnosis Analysis. ASHRAE. St. Louis, MO, 8.
[8] Singh, M. Improving building operational performance with reactive management embedding diagnosis capabilities. PhD. Thesis University of Grenoble Alpes 2017
[9] Zhang, W., Zhao, Q., Zhao, H., Zhou, G., & Feng, W. (2018). Diagnosing a Strong-Fault Model by Conflict and Consistency. Sensors, 18(4), 1016.