Abstract
In order to develop a fault detection and diagnosis (FDD) algorithm for air handling unit (AHU) based on machine learning, the economizer outdodor air temperature (OAT) sensor fault was modeled. Through EnergyPlus program, the economizer OAT sensor modeled a fault that measures the OAT that’s higher or lower than the actual temperature. The distribution of abnormal node point air temperature produced by the fault at the control point of the economizer was reviewed. In addition, the relationship between OAT & mixed air temperature (MAT), OAT & outdoor air fraction (OAF), AHU cooling coil energy, and chiller energy were comparatively analyzed for each fault model. In the case of the fault models, although outdoor air suitable for cooling could be utilized during the operation, outdoor air was not introduced for cooling, and in the opposite case, outdoor air was introduced even when OAT was higher than the indoor set-point temperature, reducing the energy saving effect for cooling. In the future, we aim to analyze energy performance and indoor air quality according to the fault in the return air temperature (RAT) sensor and error in the opening position of the damper. In addition, we plan to continue the analysis of fault data for the various elements of an AHU, and develop FDD algorithms using machine learning.
| Original language | English |
|---|---|
| Pages (from-to) | 2083-2094 |
| Number of pages | 12 |
| Journal | Journal of Mechanical Science and Technology |
| Volume | 36 |
| Issue number | 4 |
| DOIs | |
| Publication status | Published - 2022 Apr |
Bibliographical note
Publisher Copyright:© 2022, The Korean Society of Mechanical Engineers and Springer-Verlag GmbH Germany, part of Springer Nature.
Keywords
- Economizer
- EnergyPlus
- Fault detection diagnosis
- Machine learning
- Outdoor air temperature sensor
ASJC Scopus subject areas
- Mechanics of Materials
- Mechanical Engineering