水泥分解炉运行参数预测模型

为了实现水泥分解炉非线性系统的建模与控制,利于数学分析等特点的T-S型模糊推理与可以实现任意非线性映射的神经网络相结合,在水泥分解炉生产工艺分析的基础上,建立了水泥分解炉运行参数的模糊神经网络预测模型.该预测模型结合某大型水泥厂现场采集的生产数据,进行了仿真验证.结果表明,模型计算简单,模型网络运算输出值(预测值)与样本期望值(现场采集数据)相差很小,说明该模糊神经网络具有较好的预测能力和泛化能力.

Forecast Model of Cement Decomposing Furnace Performance Parameters

To obtain the modeling and control for nonlinearity system of cement decomposing furnace,T-S fuzzy inference where mathematics analysis is easy is combined with neural network which has random non-linear mapping characteristics.Based on analysis of production technology for decomposing furnace,a fuzzy neural network forecast model of decomposing furnace performance parameters is established.The model is simulated with the data collected online from a large-scale cement plant.The simulation shows that the model calculation method is simple,the output（forecast value） is close to the expectation value（data collected online）,and the fuzzy neural network has good ability of forecast and generalization.