³í¹®¸í |
¸Ó½Å·¯´× ¾Ë°í¸®ÁòÀ» ÀÌ¿ëÇÑ ÄÜÅ©¸®Æ® ½½·³ÇÁ Ç÷Π±â¹ÝÀÇ ÄÜÅ©¸®Æ® ·¹¿Ã·ÎÁö Á¤¼ö ¿¹Ãø ¸ðµ¨ °³¹ß / Development of a Concrete Rheology Parameter Prediction Model Based on Concrete Slump Flow Using a Machine Learning Algorithm |
ÀúÀÚ¸í |
ÀÌÀ¯Á¤(Yu-Jeong Lee) ; ±èÀÎÅÂ(In-Tae Kim) ; Çѵ¿¿±(Dong-Yeop Han) |
¼ö·Ï»çÇ× |
Çѱ¹ÄÜÅ©¸®Æ®ÇÐȸ³í¹®Áý, Vol.36 No.1 (2024-02) |
ÆäÀÌÁö |
½ÃÀÛÆäÀÌÁö(61) ÃÑÆäÀÌÁö(11) |
ÁÖÁ¦¾î |
·¹¿Ã·ÎÁö; ÄÜÅ©¸®Æ®; ¸Ó½Å·¯´×; Àΰø½Å°æ¸Á ; rheology; concrete; machine learning; artificial neural network |
¿ä¾à1 |
º» ¿¬±¸¿¡¼´Â ¸Ó½Å·¯´× ¾Ë°í¸®ÁòÀ» Ȱ¿ëÇÑ ±âÁ¸ÀÇ ÄÜÅ©¸®Æ® ½½·³ÇÁ Ç÷Π½ÃÇè °á°ú¸¦ ¹ÙÅÁÀ¸·Î ÄÜÅ©¸®Æ® ·¹¿Ã·ÎÁö Á¤¼ö ¿¹Ãø ¸ðµ¨À» °³¹ßÇÏ´Â °ÍÀ» ¸ñÇ¥·Î ÇÑ´Ù. º» ¿¬±¸¿¡¼´Â ¿¬±¸ ¸ñÀûÀ» ´Þ¼ºÇϱâ À§ÇØ µ¥ÀÌÅÍ Àüó¸®, ǰÁú, ÈÆ·Ã µ¥ÀÌÅÍ °³¼ö¿¡ µû¸¥ ¿¹Ãø ¸ðµ¨ÀÇ ¼º´ÉÀ» ºÐ¼®ÇÏ¿´´Ù. ºÐ¼® °á°ú, Data Cleaning°ú µ¥ÀÌÅÍ Á¤±Ôȸ¦ ¸ðµÎ Àû¿ëÇÏ´Â µ¥ÀÌÅÍ Àüó¸® ¹æ½ÄÀÌ ¸ðµ¨ÀÇ ¼º´É °³¼±¿¡ °¡Àå È¿°úÀûÀ̾ú´Ù. ¶ÇÇÑ ¿¹Ãø ¸ðµ¨ÀÇ ÇнÀ µ¥ÀÌÅÍÀÇ Ç°ÁúÀÌ ³ô°í ÈÆ·Ã µ¥ÀÌÅÍÀÇ ¼ö°¡ ¸¹À»¼ö·Ï ¿¹Ãø ¸ðµ¨ÀÇ ¼º´ÉÀÌ Çâ»óµÊÀ» È®ÀÎÇÏ¿´´Ù. º» ¿¬±¸¸¦ ÅëÇØ ±»Áö¾ÊÀº ÄÜÅ©¸®Æ®ÀÇ ½½·³ÇÁ Ç÷Πµ¥ÀÌÅ͸¦ ±â¹ÝÀ¸·Î ·¹¿Ã·ÎÁö Á¤¼ö ¿¹Ãø ¸ðµ¨À» °³¹ßÇÏ´Â µ¥ ±â¿©ÇÒ °ÍÀ¸·Î ÆÇ´ÜµÈ´Ù. |
¿ä¾à2 |
This study involved developing a predictive model for concrete rheological parameters using a machine learning algorithm, based on conventional test results of concrete slump flow. To achieve this, the prediction model¡¯s performance was assessed according to data preprocessing, data quality, and the quantity of training data. Analysis revealed that data preprocessing involving both data cleaning and normalization proved effective. Furthermore, the predictive model¡¯s performance improved with higher quality and a larger volume of training data. This study can contribute to the development of a prediction model for rheology parameters based on fresh concrete slump flow data. |