基于微分动态规划的自动烧焦系统开发及应用  

Development and Application of Automatic Control System for Decoking Process Based on Differential Dynamic Programming

作　　者：周海强 陆向东 张珍珍 林嘉斌 赵众[2] Zhou Haiqiang;Lu Xiangdong;Zhang Zhenzhen;Lin Jiabin;Zhao Zhong(Sinopec Zhenhai Refinery and Chemical Company,Ningbo,315207,China;College of Information Science and Technology,Beijing University of Chemical Technology,Beijing,100029,China)

机构地区：[1]中国石油化工股份有限公司镇海炼化分公司,浙江宁波315207 [2]北京化工大学信息科学与技术学院,北京100029

出　　处：《石油化工自动化》2025年第1期11-16,共6页Automation in Petro-chemical Industry

基　　金：2019年工业互联网创新发展工程项目(TC19084DY);北京市自然科学基金项目(4172044);北京市朝阳区协同创新项目(CYXC1707)。

摘　　要：乙烯是重要的石油化工原料,其生产周期中的烧焦过程是重要的非平稳操作过程之一,烧焦过程中关键操作参数大幅波动会影响整个装置操作的平稳性。传统的人工烧焦操作存在劳动负荷大,以及因炉管温度分布不均导致炉管堵塞或损坏的风险。基于操作经验,结合生产过程数据分析,采用微分动态规划算法(DDP)开发了自动烧焦系统,并应用于实际生产过程。现场应用表明:所开发的自动烧焦系统能平稳自动地运行烧焦操作,显著提升了大型乙烯装置非平稳操作的效率和安全性。Ethylene is an important raw material in the petrochemical industry.The decoking process in its production cycle is an important non-stationary operation where the big fluctuation of key operation parameters in decoking process could affect the stability of installation operations.Traditional manual decoking operation may have the risk of large workload and furnace tube damage or blocking due to uneven distribution of furnace tube temperature.Based on the operational experience with combination of data analysis of production process,an automatic control system for decoking process has been developed with applying the differential dynamic programming(DDP)optimization method,and is applied to the real industrial process.The application results have verified that the developed automatic decoking system can conduct the operation automatically and stably.The efficiency and safety of the large ethylene installation non-stationary operation are improved sharply.

关 键 词：乙烯裂解炉 非平稳操作 烧焦 微分动态规划 自动控制 

分 类 号：TP273[自动化与计算机技术—检测技术与自动化装置]