第一页
闪蒸计算画相态图的问题
#3
#4
#6
yolanda
发表于 02-17-2012 - 05:30
精准.石油.论坛 forum.petro-china.com
Quite several years ago, I coded for the phase envelope for many different oils/condensates. I started from a low temperature for the bubble-point and progressively use previous bubble-point pressure/temperature and it was very smooth even at .1 oF to the critical point. It will not work for sure exactly at the critical point. The calculation continues until the cricondentherm point and there is a small twist there by altering pressure afterwards instead of temperature and then calculate the reminder of the dew point curve.
In the process, the temperature step is smaller approaching the critical point until pre-set .1 oF. And there is a detector for the passing-or-not of the critical point to control the temperature steps.
There is also a flag used for detecting bubble-point curve to handle condensates.
When it approaches critical points, I remembered to switch to accelerated SS instead of Newtonian method, and you can check C.H.Whitson's book.
I forget if I did something special for the cricondenbar point or not. I have yet found a composition that the program did not handle smoothly.
精准.石油.论坛 forum.petro-china.comIn the process, the temperature step is smaller approaching the critical point until pre-set .1 oF. And there is a detector for the passing-or-not of the critical point to control the temperature steps.
There is also a flag used for detecting bubble-point curve to handle condensates.
When it approaches critical points, I remembered to switch to accelerated SS instead of Newtonian method, and you can check C.H.Whitson's book.
I forget if I did something special for the cricondenbar point or not. I have yet found a composition that the program did not handle smoothly.
引用框(JasonXu @ 11-10-2011 - 08:19)
很抱歉,一直没上论坛,大家一般都说露点线难画,因为要分段,其实就我看来,泡点线是很麻烦的,因为牵涉到的最优解震荡比较大,而其中的细节很多论文没有提及,最近实现了这个工作,但是临界点附近仍然震荡非常明显,不过现有软件基本都是这个情况
----
来源: 精准石油论坛 - 推进信息共享,提升科技水平
第一页

登录
注册
帮助


多重回复