结合现场数据，采用理论与经验、静态与动态相结合的方法，开发了鲁姆斯ＳＲＴ－ Ⅲ型柴油裂解炉动态仿真模型。辐射管内物料采用由神经网络产生的拟静态模型，管外及炉膛的模型中引入了部分经验参数。仿真结果表明本文开发的模型适用范围广、逼真度高，满足动态模型适时性和精度要求。 Combined with the field data, the dynamic simulation model of SRS-Ⅲ diesel cracking furnace was developed by combining theory and experience, static and dynamic methods. The material in the radiant tube adopts the quasi-static model generated by the neural network, and some empirical parameters are introduced into the model outside the tube and the furnace. The simulation results show that the model developed in this paper is applicable to a wide range of applications with high fidelity to meet the requirements of the dynamic model in timeliness and accuracy.