STUDY ON PREDICT THE TEMPERATURE OF READY-MIXED CONCRETE IN OKINAWA PREFECTURE BY RANDOM FOREST

  • SHIMIZU Kanta
    UNIVERSITY OF The RYUKYUS, Grad. School of Eng. and Sci.(1, Sembaru, Nishihara-cho, Nakagami-gun, Okinawa 903-0213, Japan)
  • YAMADA Yoshitomo
    UNIVERSITY OF THE RYUKYUS, Faculty of Eng.(1, Sembaru, Nishihara-cho, Nakagami-gun, Okinawa 903-0213, Japan)
  • KOYAMA Tomoyuki
    KYUSYU UNIVERSITY, Grad. School of Human-Environment Studies(744, Motooka, Nishi-ku, Fukuoka-shi, Fukuoka 819-0395, Japan)

Bibliographic Information

Other Title
  • ランダムフォレストによる沖縄県の生コンクリート温度予測の試み

Abstract

<p>In this study, concrete temperature shipped in Okinawa Prefecture was trained using a random forest (RF), which is a type of machine learning, from the features (explanatory variables) such as materials used, formulation, outside temperature, total solar radiation, and weather. Using this RF learning, we tried to predict the concrete temperature at the time of kneading and unloading. In addition, the importance of each feature obtained from the learning was evaluated, and multiple regression analysis was performed using the highly important features. Comparing the proposed multiple regression equation with the equations proposed by the Architectural Institute of Japan and the Japan Society of Civil Engineers, it was confirmed that the proposed multiple regression equation can predict the concrete temperature more accurately.</p>

Journal

References(1)*help

See more

Details 詳細情報について

Report a problem

Back to top