• Àüü
  • ÀüÀÚ/Àü±â
  • Åë½Å
  • ÄÄÇ»ÅÍ
´Ý±â

»çÀÌÆ®¸Ê

Loading..

Please wait....

±¹³» ÇÐȸÁö

Ȩ Ȩ > ¿¬±¸¹®Çå > ±¹³» ÇÐȸÁö > µ¥ÀÌÅͺ£À̽º ¿¬±¸È¸Áö(SIGDB)

µ¥ÀÌÅͺ£À̽º ¿¬±¸È¸Áö(SIGDB)

Current Result Document : 4 / 78 ÀÌÀü°Ç ÀÌÀü°Ç   ´ÙÀ½°Ç ´ÙÀ½°Ç

ÇѱÛÁ¦¸ñ(Korean Title) ÆÄƼŬ ÇÊÅÍ ±â¹Ý ½Ã°è¿­ ¿¹Ãø °áÇÕ ¸ðµ¨À» ÅëÇÑ Àü½ÃÀå ¹æ¹® Àοø ¿¹Ãø
¿µ¹®Á¦¸ñ(English Title) Predicting the number of exhibition visitors using a time-series prediction model combined with particle filter
ÀúÀÚ(Author) °­¼®¿ì   ±è¼ºÇö   ±ÇÁØÈ£   ¼Û±æÅ   Seokwoo Kang   Seonghyeon Kim   Joonho Kwon   Giltae Song  
¿ø¹®¼ö·Ïó(Citation) VOL 34 NO. 01 PP. 0003 ~ 0013 (2018. 04)
Çѱ۳»¿ë
(Korean Abstract)
ÄèÀûÇÑ È¯°æÀ» Á¦°øÇϸ鼭 ¿¡³ÊÁö »ç¿ë·®À» ÃÖÀûÈ­ÇÏ´Â ±â¹ý¿¡ ´ëÇÑ ¿¬±¸´Â Àü½ÃÀå °ü¸® ½Ã½ºÅÛÀÇ ÁÖ¿ä ¹®Á¦ Áß ÇϳªÀÌ´Ù. ÄèÀûÇÑ Àü½Ã ȯ°æÀ» À¯ÁöÇÔ°ú µ¿½Ã¿¡ Àü½ÃÀå ¿¡³ÊÁö »ç¿ë·®À» ÃÖÀûÈ­Çϱâ À§Çؼ­ Àç½Ç Àοøµ¥ÀÌÅÍ ºÐ¼®ÀÌ ÇʼöÀûÀÌ´Ù. Àü½ÃÀå¿¡¼­ ¼öÁýµÈ ½Ã°£º° ¹æ¹®°´ µ¥ÀÌÅÍ ºÐ¼®¿¡ ±â¹Ý ÇÑ ¹Ì·¡ Àç½Ç Àοø ¿¹ÃøÀ» ÅëÇØ ÃÖÀûÀÇ ¿¡³ÊÁö¸¦ »ç¿ëÇÏ´Â Àü½ÃÀå ³Ã³­¹æ Á¦¾î °èȹÀ» ¹Ì¸® ¼ö¸³ÇÏ°í À̸¦ Àü½ÃÀå Á¦¾î ½Ã½ºÅÛ¿¡ Àû¿ëÇÒ ¼ö ÀÖ´Ù. Á¤È®ÇÑ ¹Ì·¡ Àç½Ç Àοø ¿¹ÃøÀ» À§ÇÏ¿© ½Ã°è¿­ ¿¹Ãø¿¡ ³Î¸® »ç¿ëµÇ´Â Holt-Winters ¸ðÇü°ú ARIMA ¸ðÇüÀ» Àû¿ëÇØ º¼ ¼ö ÀÖÀ¸³ª ¿¹Ãø ½ÃÁ¡ÀÌ ¸Ö¾îÁú¼ö·Ï ¿¹Ãø Á¤È®µµ°¡ ¶³¾îÁø´Ù´Â ¹®Á¦Á¡ÀÌ ÀÖ´Ù. º» ¿¬±¸¿¡¼­´Â µÎ ´ëÇ¥ÀûÀÎ ½Ã°è¿­ ¸ðÇüÀ» È°¿ëÇÏ¿© ¿ì¼± ½Ã°£º° Àοø µ¥ÀÌÅÍ ¿¹ÃøÀ» ÁøÇàÇÏ°í ¸Ó½Å·¯´×ÀÇ ±â¹ýÀÇ ÇϳªÀÎ ÆÄƼŬ ÇÊÅ͸¦ ÀÌ¿ëÇÑ ¿ÀÂ÷ ½º¹«µùÀ¸·Î µÎ ¸ðµ¨ÀÇ °á°ú¸¦ °áÇÕÇÏ´Â ½Ã°è¿­ ¿¹Ãø ¸ðµ¨À» Á¦¾ÈÇÑ´Ù. Á¦¾ÈÇÑ ¸ðµ¨À» ŲÅؽº Á¦ 7Ȧ Àü½ÃÀå¿¡¼­ ÁøÇàµÈ 2017³â Àü½Ã °ü¶÷°´ µ¥ÀÌÅÍ¿¡ Àû¿ëÇÏ¿© ±âÁ¸ÀÇ ¿¹Ãø ¸ðµ¨°úÀÇ Á¤È®µµ ºñ±³¸¦ ÅëÇØ Á¦¾ÈÇÑ ¸ðµ¨À» °ËÁõÇÏ¿´´Ù. ¹Ì·¡ Àç½Ç Àοø¿¡ ´ëÇÑ Á¤È®ÇÑ ¿¹ÃøÀ¸·Î º¸´Ù È¿À²ÀûÀÎ ´ëÇü Àü½ÃÀå ³Ã³­¹æ Á¦¾î¸¦ ÅëÇÑ ¿¡³ÊÁö »ç¿ë ÃÖÀûÈ­¸¦ ´õ¿í ¾Õ´ç±æ °ÍÀ¸·Î ±â´ëÇÑ´Ù.
¿µ¹®³»¿ë
(English Abstract)
The demand for exhibition building management has been focused on optimizing energy consumption while providing a pleasant environment. To investigate a method for optimizing energy usage while maintaining the good exhibition condition, the analysis of visitors¡¯ occupancy data is essential. The accurate prediction of visitors¡¯ occupancy enables to schedule an efficient air conditioning plan and to apply it to air conditioning control systems. Holt-Winters and ARIMA models popular in time series data analysis can be applied to predict the occupancy of visitors accurately, but the prediction accuracy of these models drops substantially as the prediction time-point becomes further. In this study, we applied Holt-Winters and ARIMA models for predicting the number of visitors in various time-points first and then combined two results using error smoothing via a machine learning approach, particle filter. The prediction results obtained by our proposed model was compared with the pure ARIMA and Holts-Winter models for evaluation using visitors¡¯ occupancy data collected in real time at the KINTEX exhibition hall 7 in 2017. The accurate prediction of building occupancy will accelerate the optimization of energy consumption via more efficient air conditioning control in large scale exhibition halls.
Å°¿öµå(Keyword) Àç½ÇÀοø µ¥ÀÌÅÍ   ½Ã°è¿­ ¿¹Ãø   ARIMA   Holt-Winters   ÆÄƼŬ ÇÊÅÍ   Building occupancy data   Time-series prediction   ARIMA   Holt-Winters   Particle filter  
ÆÄÀÏ÷ºÎ PDF ´Ù¿î·Îµå