Posted Tue, 14 May 2024 13:56:22 GMT by Simon Freitag
Dear Feflow-community, I'm quite new to working with feflow and hence, am struggling with some issues. I've built a quite simple structured 3D-model with several layers including an aquifer that is confined by low-permeable layers. This aquifer should serve as an ATES-aquifer with several injection and production wells. All wells are already implemented with the ideal cell size, according flow rates as well as a temperature bc at the bottom node of each well. First simulations are working quite well, however, I'm observing a weird effect, as after the first year (1 injection and 1 production cycle), I can extract more energy/heat than I injected, although I injected 90°C into a reservoir with 13°C ambient temperature. Has anyone observed a similar effect and explain the reason? I am seeing another effect I cannot explain, as after each switching point (from injection to quiet phase and from quiet phase to production), the temperature shortly drops by ~20°C...why? What's the problem here? And one last question: If I group the wells and want to assign a temperature difference in the "Open Loop" property of this well group, how do I need to prepare the t-difference time series/how do I use this option? What t-difference is meant here (I don't really understand the feflow documentation explanation)? Can anyone maybe help me with these three issues? Many thanks, Simon
Posted Mon, 17 Jun 2024 15:11:46 GMT by Alexander Magerl Joswig
Hi Simon,  about your last question, there is a Video [https://www.youtube.com/watch?v=MGvhyww_hKk] describing the steps necessary to implement an open loop System using the OpenLoop plugin. I think starting with Feflow 8.1 the function is included in the GUI and the workflow might differ.  Cheers  Alex   
Posted Tue, 18 Jun 2024 07:04:55 GMT by Carlos Andres Rivera Villarreyes Global Product Specialist - FEFLOW
Yes, with the latest FEFLOW 8.1, we strongly do not recommend to use the Open-Loop plug-in. The full functionality is implemented in the GUI. This will also warranty a robust model performance. Cheers Carlos

You must be signed in to post in this forum.