主催: 日本エネルギー学会, 石炭科学部会, コークス工学研究部会, 重質油部会
共催: 石炭・炭素資源有効利用研究会
会議名: 第59回石炭科学会議
開催地: 札幌・道民活動センターかでる2.7（ハイブリッド開催）
開催日: 2022/10/20 - 2022/10/21
p. 52-53
Stable operation of coal gasifier is the key subject for practical use of Integrated coal Gasification Combined Cycle (IGCC). It is necessary to satisfy the condition of keeping stable slag discharge with holding temperature in a gasifier higher than fusion temperature of the slag. This temperature for gasifier’s operation has been determined on the basis of the experimental coal ash fusion temperature, which requires a lot of work and time. Therefore, in order to improve the operability of IGCC, coal ash fusion temperatures were predicted using chemical composition of ash and machine learning technique in this study. As a result, it was confirmed that the most accurate predicted value can be obtained by creating a model that combines outlier removal by LOF and regression analysis by Random Forest.
