| ³í¹®¸í |
VAV ½Ã½ºÅÛÀÇ µ¥ÀÌÅÍ ÇÊÅ͸µ°ú ¿À·ù °ËÃâ / Data Filtering and Fault Detection of VAV System / 2-4 : ºôµù½Ã¹Ä·¹ÀÌ¼Ç I |
| ÀúÀÚ¸í |
±è¿µÁø(Kim, Young-Jin) ; ¾È±â¾ð(Ahn, Ki-Uhn) ; ±è±âö(Kim, Ki-Cheol) ; ¹Úö¼ö(Park, Cheol-Soo) |
| ¼ö·Ï»çÇ× |
Ãß°èÇмú¹ßÇ¥´ëȸ, 2015 (2015-11) |
| ÆäÀÌÁö |
½ÃÀÛÆäÀÌÁö(165) ÃÑÆäÀÌÁö(2) |
| ÁÖÁ¦¾î |
¿À·ù °ËÃâ ; µ¥ÀÌÅÍ ÇÊÅ͸µ ; °Ç¹° ¿¡³ÊÁö ¿î¿µ ½Ã½ºÅÛ ; º¯Ç³·®°øÁ¶ ; ºôµù ½Ã¹Ä·¹ÀÌ¼Ç ; Fault detection ; Data filtering ; BEMS ; VAV ; Building Simulation |
| ¿ä¾à2 |
For efficient operations of HVAC systems under Building Energy Management System (BEMS) environment, sensor data filtering and fault detection methods are needed. This paper addresses an automating data filtering and fault detection model of a Variable Air Volume (VAV) system where discrete wavelet transform and data driven machine learning model are coupled. To validate the developed model, three fault scenarios (temperature and mass flow rate sensor offset, outdoor air damper stuck or leaking, and coil valve control fault) are chosen and tested. In the paper, it was found that the data filtering and fault detection model provide building operators with meaningful system fault alarms. |