基于工业网络的电弧炉神经网络控制系统

摘要

介绍了基于工业计算机网络的电弧炉控制系统,该系统采用三层网络结构即管理层、控制层、设备层完成电弧炉和精练炉的控制,实现了炼钢过程的优化控制及数据监控,完成了工艺优化,提高了生产效率。电弧炉和精练炉控制采用基于三相意识的专家控制系统,并提出一种新的电弧炉控制算法预估补偿算法。通过预估对象的状态,对常规控制器的输出做补偿控制,使系统较好地适应负荷变化和外部的干扰,使电弧炉控制过程性能大大提高。%This paper presents an electric furnace control system consisting of Ethernet, control net and device net.The system achieves the optimized control and the data monitor of steelmaking process, operation optimization,with which production efficiency has been improved.An expert system is adapted in the system.The paper also presents a predictive compensation algorithm based on the neural network;the algorithm is used to predict the condition of E.A.F,by which the output of a routine controller can be controlled.The system can adapt to load variation and external interruption,yielding control quality.