シャ　キンカ   Jinhua She   佘　錦華    所属   工学部 機械工学科    職種   教授
言語種別	英語
発行・発表の年月	2017/07
形態種別	国際会議論文
査読	査読あり
標題	Hybrid Prediction Model of Carbon Efficiency for Sintering Process
執筆形態	共著
掲載誌名	The 20th World Congress of the International Federation of Automatic Control (IFAC 2017)
掲載区分	国外
出版社・発行元	The International Federation of Automatic Control
巻・号・頁	pp.10660-10665
担当範囲	ALL
著者・共著者	Jie Hu; Min Wu, Xin Chen, Jinhua She, Weihua Cao, Luefeng Chen, and Huafeng Ding
概要	The main source of energy consumption is the consumption of carbon. It is significant to predict the carbon efficiency to reduce the energy consumption. In the paper, first, the comprehensive coke ratio (CCR) is defined and used as an index to measure the carbon efficiency by analyzing the sintering mechanism. The principal component analysis method is used to find the principal components affecting the CCR. Next, they are divided into different subclasses by adopting the fuzzy C-means clustering algorithm. Then, the least square-support vector machine (LS-SVM) sub-models are established based on the subclasses, and the parameters of the sub-models are obtained using an adaptive particle swarm optimization algorithm. Finally, a FCM-LSSVM model is established by weighting each LS-SVM sub-model with a fuzzy membership function. The simulations using actual production data show that the prediction accuracy of the FCM-LSSVM model is higher than that of a BP neural network model and a single LS-SVM model, and it meets the requirements of actual production.
外部リンクURL	https://www.sciencedirect.com/science/article/pii/S2405896317317974