【作者】 吴永建； 张晶； 张莉； 岳恒； 柴天佑；

【Author】 Wu YongJian~1 Zhang Jing~1 Zhang Li~(1,2) Yue Heng~(1,2) Chai Tianyou~(1,2) (1 Key Laboratory of Process Industry Automation Ministry of Education Shenyang 1100040 China;2 Research Center of Automation Northeastern University Shenyang 110004 China;3 P&G Manufacturing(Tianjin) Co LtdTianjin 300385 China)

【机构】 东北大学暨流程工业综合自动化教育部重点实验室； 东北大学自动化研究中心； 天津宝洁工业有限公司；

【摘要】 针对电熔镁炉强耦合、非线性、难以建立精确数学模型的特点,把机理建模和人工智能辨识方法相结合,进行了电熔镁炉电极调节系统建模研究和仿真验证。紧密结合冶炼工艺,通过对电熔镁炉电极调节系统各组成部分工作机理的分析,在依据工业电炉相关电热理论知识进行合理假设和简化的基础上,建立了电熔镁炉电极调节系统的机理模型:为弥补机理模型忽略外界扰动、工业应用精度低等缺陷,在现场实测数据的支持下建立了基于Elman回归神经网络模型,对机理模型的输出进行动态补偿,仿真结果证明了混合模型逼近精度高。更多还原

【Abstract】 Since the electrical smelting furnace for magnesia is an object with strong nonlinearity time variation and strong coupling it is difficult to build its mathematics model A hybrid modeling methodology integrating the mechanism modeling and artificial intelligent identification method was proposed and the modeling of electrode regulating system of the electrical smelting furnace for magnesia was studied and simulated The mechanism model of the electrodes adjustment system was built through the analysis of working principle of every component of electrodes adjustment systems combining with the smelting techniques rational hypothesis and simplification based on the relational electric-thermal theory In order to compensate the bugs such as ignoring outside disturbs low precision and so on Elman regression neural network model was built in support of field data The mechanism model can be revised by the output of Elman regression neural network model timely.The simulation shows that the hybrid model is of high approaching accuracy.更多还原