回转窑的生产过程是一个复杂的物理化学反应过程，具有大惯性、纯滞后、非线性等

特点。工艺过程复杂多变，难以得到精确的数学模型.本文利用BP神经网络可以实现任意非

线性映射的特点建立其神经网络预测模型,结合广西某大型水泥厂实时采集的生产数据，进

行仿真研究。仿真结果表明，该模型能够很好的预测水泥回转窑的温度。 -Rotary kiln production process is a complex physical and chemical reaction process, with a large inertia, pure lag and nonlinear characteristics. Complex process, difficult to get a precise mathematical model. In this paper, BP neural network can implement any nonlinear mapping to establish the characteristics of the neural network prediction model, combined with a large cement plant in Guangxi real-time collection of production data, the simulation study. Simulation results show that the model can well predict the temperature of the cement kiln.