基于BP神经网络和CFD数值模拟的气旋浮罐结构优化及性能预测

发布时间：2019-06-21 08:33

【摘要】：采用CFD模拟和BP神经网络结合的方法,对BIPTCFU-III型气旋浮装置主体设备气旋浮罐的入口管径、环形缝隙宽度、高径比和稳流筒直径等主要结构参数进行了优化,研究了结构参数对分离效率的影响.结果表明,优化后气旋浮罐分离效率从81.07%提升至92.82%,预测值与模拟值的偏差仅为1.15%;增大稳流筒直径、提高高径比、增大入口管径和减小环形缝隙宽度有利于强化油水两相的运移和分离过程,提高分离效率.
[Abstract]:The main structural parameters such as the inlet pipe diameter, the annular gap width, the high-diameter ratio and the diameter of the steady-flow tube were optimized by using the method of CFD simulation and BP neural network, and the effect of the structural parameters on the separation efficiency was studied. The results show that the separation efficiency of the post-cyclone floating tank is increased from 81.07% to 92.82%, the deviation between the predicted value and the simulated value is only 1.15%, the diameter of the steady flow cylinder is increased, the high-diameter ratio is increased, the inlet pipe diameter and the width of the annular gap are increased, and the migration and separation process of the oil-water two-phase is facilitated. And the separation efficiency is improved.
【作者单位】：北京化工大学机电工程学院;北京石油化工学院机械工程学院;
【基金】：北京市属高等学校“长城学者”培养计划资助项目(编号:CIT&TCD20150317) 北京市自然科学基金资助项目(编号:KZ201410017019)
【分类号】：X741

【相似文献】