延迟焦化装置碳钢加热炉管道损坏预测——基于神经网络的局部温度实时检测方法  

Prediction of Pipeline Damage in Carbon Steel Heating Furnace of Delayed Coking Plant-A Real Time Local Temperature Detection Method Based on Neural Network

作　　者：李朝阳 Li Chaoyang(Shanxin Software Rizhao Branch,Rizhao Shandong 276800,China)

机构地区：[1]山信软件日照分公司,山东日照276800

出　　处：《山西冶金》2023年第8期125-126,132,共3页Shanxi Metallurgy

摘　　要：在延迟焦化装置中,碳钢加热炉存在燃烧不稳定性,易造成加热炉管内局部超温而损坏的问题,因此必须实时检测加热炉内各个部位的温度。针对这一问题,拟开展基于卷积长短时记忆神经网络的炉温在线预报方法研究,通过对高温炉温变化规律的分析,实现对高温炉温变化的预测,并将平均误差控制在31.5 Kalvin以下。In delayed coking equipment,carbon steel heating furnaces have combustion instability,which can easily cause local overheating and damage in the heating furnace tubes.Therefore,it is necessary to monitor the temperature of various parts of the heating furnace in real-time.In response to this issue,it is planned to conduct research on the online prediction method of furnace temperature based on convolutional long and short time memory neural networks.By analyzing the changes in high-temperature furnace temperature,the prediction of high-temperature furnace temperature changes is achieved,and the average error is controlled below 31.5 Kalvin.

关 键 词：加热炉管道 神经网络 实时温度 损坏检测 

分 类 号：TE96[石油与天然气工程—石油机械设备]