CAJ | 학술논문

为了解决在炉长操作炼铁制度及产量至上的原则下，能源和资源长期处于“盲目”消耗的问题，利用含有高炉冶炼过程客观规律和生产技术人员操作决策等重要信息的历史生产数据，运用数据挖掘中的模式匹配方法在喷煤操作模式库中匹配得到当前喷煤量，若未能匹配到则采用BP神经网络和粒子群寻优算法进行演化，通过提高喷煤量降低焦比从而达到节能降耗的目的。
위료해결재로장조작련철제도급산량지상적원칙하，능원화자원장기처우“맹목”소모적문제，이용함유고로야련과정객관규률화생산기술인원조작결책등중요신식적역사생산수거，운용수거알굴중적모식필배방법재분매조작모식고중필배득도당전분매량，약미능필배도칙채용BP신경망락화입자군심우산법진행연화，통과제고분매량강저초비종이체도절능강모적목적。
The energy and resource consumption is in a“blind” state over a long period of time un-der the ironmaking system of the furnace superintendent operation and the principle of the primacy of yields. In order to solve the problem,we take advantage of the historical data containing important in-formation such as the objective laws of the blast furnace smelting process and the operating decisions of production technicians,and use the pattern matching method of data mining to get the current coal injection rate through matching in the library of coal injection operation patterns. If the matching fails,the BP neural network and particle swarm optimization algorithm are then used to evolve the op-eration pattern to achieve the purpose of saving energy and reducing consumption by raising coal in-jection rate and decreasing coke ratio.