渣油加氢装置加热炉结焦量化表征及预测  

Quantitative characterization and prediction for coking in furnace of residual oil hydrogenation unit

作　　者：张兵 郑晓军[1] 杨彩娟[1] 张峥 付纪林 ZHANG Bing;ZHENG Xiao-jun;YANG Cai-juan;ZHANG Zheng;FU Ji-lin(Petro-Cyber Works Information Technology Co Ltd,Shanghai 200040,China)

机构地区：[1]石化盈科信息技术有限责任公司,上海200040

出　　处：《石化技术与应用》2023年第5期363-366,376,共5页Petrochemical Technology & Application

基　　金：中国石化科技开发基金资助项目(项目编号:322107&320130)。

摘　　要：为了准确且快速判断渣油加氢装置加热炉结焦程度、预判结焦趋势,从能量传递角度设计出结焦量化表征指标(N),N为实际操作条件且炉管在洁净状态下的内传热系数除以其在同等操作条件及结焦状态下的内传热系数所得值;采用Aspen HYSYS软件模拟建立了N值变化预测模型,并结合某公司300万t/a渣油加氢装置生产数据进行了验证。结果表明:经计算所得N值及其变化趋势与实际生产运行情况满足一致性,皮尔逊相关度计算结果满足排他性要求;所建立的结焦量化预测模型的准确性较好,历史同期的实际N值与其回归值的自检相对误差均值为3.15%;在跟踪同装置生产运行的4个月期间,预测模型的实际N值与预测值的变化总体趋势基本一致,相对预报误差均值为3.90%。In order to accurately and quickly determine the coking degree and predict coking trend in the furnace of residual oil hydrogenation unit,the quantitative index of coking(N)was designed according to energy transfer,which was the internal heat transfer coefficient without coking divided by the internal heat transfer coefficient with coking under the same operation condition.The prediction model for N was established by Aspen HYSYS software.The model was confirmed by production data from a 3 Mt/a residual oil hydrogenation unit.The results showed that the calculated N value and its trend of change was consistent with the actual condition,and the Person correlation coefficient met the exclusivity requirement.The established coking quantitative prediction model had good accuracy.The average self-checking relative error between the actual N value and its regression value during the same period in history was 3.15%.During next 4 months period of tracking production and operation of the unit,the overall trend of the predicted N value was basically consistent with the actual N value,with the average relative prediction error of 3.90%.

关 键 词：渣油加氢装置 加热炉炉管 结焦 量化表征 预测 

分 类 号：TQ021.8[化学工程]