¶óÆæÆ®¦¢Ä«Æä¦¢ºí·Î±×¦¢´õº¸±â
¾ÆÄ«µ¥¹Ì Ȩ ¸í»çƯ°­ ´ëÇבּ¸½Ç޹æ Á¶°æ½Ç¹« µ¿¿µ»ó°­ÀÇ Çѱ¹ÀÇ ÀüÅëÁ¤¿ø ÇÐȸº° ³í¹®
ÇÐȸº° ³í¹®

Çѱ¹°Ç¼³°ü¸®ÇÐȸ
Çѱ¹°ÇÃà½Ã°øÇÐȸ
Çѱ¹µµ·ÎÇÐȸ
Çѱ¹»ý¹°È¯°æÁ¶ÀýÇÐȸ
Çѱ¹»ýÅÂÇÐȸ
Çѱ¹¼öÀÚ¿øÇÐȸ
Çѱ¹½Ä¹°ÇÐȸ
Çѱ¹½Ç³»µðÀÚÀÎÇÐȸ
Çѱ¹ÀÚ¿ø½Ä¹°ÇÐȸ
Çѱ¹ÀܵðÇÐȸ
Çѱ¹Á¶°æÇÐȸ
Çѱ¹Áö¹Ý°øÇÐȸ
Çѱ¹ÇÏõȣ¼öÇÐȸ
Çѱ¹È¯°æ»ý¹°ÇÐȸ
Çѱ¹È¯°æ»ýÅÂÇÐȸ

Çѱ¹¼öÀÚ¿øÇÐȸ / v.33, no.5, 2000³â, pp.537-550
´ÙÃþ½Å°æ¸Á¸ðÇü¿¡ ÀÇÇÑ ÀÏ À¯Ãâ·®ÀÇ ¿¹Ãø¿¡ °üÇÑ ¿¬±¸
( A Study on the Forecasting of Daily Streamflow using the Multilayer Neural Networks Model )
±è¼º¿ø; ÄÝ·Î¶óµµ ÁÖ¸³´ëÇб³ Åä¸ñ°øÇаú;
 
ÃÊ ·Ï
º» ¿¬±¸¿¡¼­´Â ³«µ¿°­ Áøµ¿ÁöÁ¡¿¡¼­ ÀÏÀ¯Ãâ·®À» ¿¹ÃøÇϱâ À§ÇÏ¿© ½Å°æ¸Á¸ðÇüÀÌ Á¦½ÃµÇ¾ú´Ù. ½Å°æ¸Á¸ðÇüÀÇ ±¸Á¶´Â CASE 1(5-5-1)°ú CASE 2(5-5-5-1)·Î ±¸¼ºÇÏ¿´À¸¸ç, Àº´ÐÃþÀÇ ¼ö¿¡ µû¶ó µÎ °¡ÁöÀÇ ¸ðÇüÀ¸·Î ºÐ·ùÇÏ¿´´Ù. °¢ ½Å°æ¸Á¸ðÇüÀº ±¤¿ªÃÖ¼ÒÁ¡°ú ÈÆ·ÃÀÓ°èÄ¡¿¡ ¼ö·ÅÇϴµ¥ ±âÁ¸ÀÇ ¿ªÀüÆÄÈÆ·Ã ¾Ë°í¸®Áò(BP) º¸´Ù ¶Ù¾î³­ Fletcher-Reeves °ø¾×±¸¹è ¿ªÀüÆÄÈÆ·Ã ¾Ë°í¸®Áò(FR-CGBP)°ú ÃàÀûµÈ °ø¾×±¸¹è ¿ªÀüÆÄÈÆ·Ã ¾Ë°í¸®Áò(SCGBP)À» ÀÌ¿ëÇÏ¿´´Ù. ±×¸®°í ¸ðÇüÀÇ ÈÆ·Ã°ú °ËÁõÀ» À§ÇÏ¿© ÀÌ¿ëµÈ ÀÚ·á´Â dz¼ö³â, Æò¼ö³â, °¥¼ö³â dz¼ö³â+Æò¼ö³â, dz¼ö³â+°¥¼ö³â, Æò¼ö³â+°¥¼ö³â ¹× dz¼ö³â+Æò¼ö³â+°¥¼ö³âÀ¸·Î ±¸ºÐÇÏ¿© ±¸¼ºÇÏ¿´´Ù. ¸ðÇüÀÇ ÈÆ·Ã°úÁ¤¿¡¼­ °¢ ÀڷḦ ÀÌ¿ëÇÏ¿© ÃÖÀû ¿¬°á°­µµ¿Í ÆíÂ÷°¡ °áÁ¤µÇ¾î Á³À¸¸ç, µ¿½Ã¿¡ ÀÏÀ¯Ãâ·®ÀÌ °è»êµÇ¾îÁ³´Ù. ¿¹Ãø¿ÀÂ÷ÀÇ Åë°èºÐ¼®À» ÅëÇÏ¿© dz¼ö³â+°¥¼ö³âÀÇ ÀڷḦ Á¦¿ÜÇϰí´Â ÈÆ·Ã°á°ú°¡ ¾çÈ£ÇÑ °ÍÀ¸·Î ³ªÅ¸³µ´Ù. ¸ðÇüÀÇ °ËÁõ¿¡´Â ¸ðÇüÀÇ ÈÆ·ÃÀ» ÅëÇØ »êÁ¤µÈ CASE 1 ÀÇ SCGBP ¾Ë°í¸®ÁòÀÇ ¿¬°á°­µµ¿Í ÆíÂ÷¸¦ ÀÌ¿ëÇÏ¿´À¸¸ç, °ËÁõÀÇ °á°ú´Â ÈÆ·Ã°á°úó·³ ¸¸Á·½º·¯¿î °ÍÀ¸·Î ºÐ¼®µÇ¾ú´Ù. ¶ÇÇÑ º» ¿¬±¸¿¡¼­ ¼±Á¤ÇÑ ½Å°æ¸Á¸ðÇü°ú ºñ±³°ËÅäÇϱâ À§ÇÏ¿© ´ÙÁßȸ±ÍºÐ¼®¸ðÇüÀ» Àû¿ëÇÏ¿© ÀÏÀ¯Ãâ·®À» ¿¹ÃøÇÏ¿´À¸¸ç, ±× °á°ú ½Å°æ¸Á¸ðÇüÀÌ ´Ù¼Ò ¿ì¼öÇÑ °á°ú¸¦ ³ªÅ¸³»´Â °ÍÀ¸·Î ºÐ¼®µÇ¾ú´Ù. ÀÌ¿Í °°ÀÌ ½Å°æ¸Á¸ðÇüÀº Á¶Á÷ÀûÀÎ Á¢±Ù¹ý, ¸Å°³º¯¼öÀÇ °¨¼Ò ¹× ¸ðµ¨À» °³¹ßÇϴµ¥ ¼Ò¸ðµÇ´Â ½Ã°£À» ÁÙÀϼö ÀÖ´Â ÀåÁ¡ÀÌ ÀÖ´Ù.
In this study, Neural Networks models were used to forecast daily streamflow at Jindong station of the Nakdong River basin. Neural Networks models consist of CASE 1(5-5-1) and CASE 2(5-5-5-1). The criteria which separates two models is the number of hidden layers. Each model has Fletcher-Reeves Conjugate Gradient BackPropagation(FR-CGBP) and Scaled Conjugate Gradient BackPropagation(SCGBP) algorithms, which are better than original BackPropagation(BP) in convergence of global error and training tolerance. The data which are available for model training and validation were composed of wet, average, dry, wet+average, wet+dry, average+dry and wet+average+dry year respectively. During model training, the optimal connection weights and biases were determined using each data set and the daily streamflow was calculated at the same time. Except for wet+dry year, the results of training were good conditions by statistical analysis of forecast errors. And, model validation was carried out using the connection weights and biases which were calculated from model training. The results of validation were satisfactory like those of training. Daily streamflow forecasting using Neural Networks models were compared with those forecasted by Multiple Regression Analysis Mode(MRAM). Neural Networks models were displayed slightly better results than MRAM in this study. Thus, Neural Networks models have much advantage to provide a more sysmatic approach, reduce model parameters, and shorten the time spent in the model development.
 
Ű¿öµå
½Å°æ¸Á¸ðÇü;´ÙÁßȸ±ÍºÐ¼®¸ðÇü;¿¹Ãø;FR-CGBP;SCGBP;Neural Networks model;Multiple Regession Analysis Model;forecasting;
 
Çѱ¹¼öÀÚ¿øÇÐȸ³í¹®Áý / v.33, no.5, 2000³â, pp.537-550
Çѱ¹¼öÀÚ¿øÇÐȸ
ISSN : 1226-6280
UCI : G100:I100-KOI(KISTI1.1003/JNL.JAKO200011920729954)
¾ð¾î : Çѱ¹¾î
³í¹® Á¦°ø : KISTI Çѱ¹°úÇбâ¼úÁ¤º¸¿¬±¸¿ø
¸ñ·Ïº¸±â
ȸ»ç¼Ò°³ ±¤°í¾È³» ÀÌ¿ë¾à°ü °³ÀÎÁ¤º¸Ãë±Þ¹æÄ§ Ã¥ÀÓÀÇ ÇѰè¿Í ¹ýÀû°íÁö À̸ÞÀÏÁÖ¼Ò ¹«´Ü¼öÁý °ÅºÎ °í°´¼¾ÅÍ
   

ÇÏÀ§¹è³ÊÀ̵¿