2 of 39

Thermo-hydraulic performance analysis of concentric pipes for geo-thermal application

N Anand, F Tosto, C de Servi, J van Bael.

Energy Systems and Components Optimization (ESCO),

Belgium.

This project has received funding from the European Union’s Horizon Europe research and innovation program under grant agreement No. 101083558 (HOCLOOP).

SU2 Conference 2024

vito.be

3 of 39

Background

Private & Confidential

4 of 39

Geothermal Power plant

vito.be

5 of 39

Schematic

Cold Fluid

Ground

Image Courtesy: Reelwell

insulation

Schematic Representation: VITO

Hot Fluid

vito.be

6 of 39

Objectives

To analyze the affect of eccentricity on the thermo-hydraulic performance of the pipe.

Investigating alternative ways to increase heat transfer in the pipe.

C = 0%

C = 100%

Casing outer

Inner pipe outer

Vacuum outer

vito.be

7 of 39

Methodology

Private & Confidential

8 of 39

Schematic

Image Courtesy: Reelwell

Hot Fluid

Cold Fluid

Ground

insulation

Schematic Representation: VITO

vito.be

9 of 39

Streamwise Periodic Flow Solver

Reduced-order Modeling

vito.be

10 of 39

SU2

Master

su2code/SU2 (V 8.0.0)

Flow Source Terms: Tobias Kattmann*

* Original contributor

Current scope limited to cold-fluid only,

SP: Streamwise Periodic

vito.be

11 of 39

Computational Setup

Options	Selections
Solver	Incompressible
Equ.	RANS+SWP
Turb.	SST-m2003v
Thermophysical Properties	Constant density
Avg. y+	5

SWP: Streamwise Periodic

Surface	BCs
Inflow	Periodic inflow (SWP)
Outflow	Periodic outflow (SWP)
Outer Dia.	Heat flux = Specified
Inner Dia.	Heat flux = 0 (insulated)

vito.be

12 of 39

Flow Domain

vito.be

13 of 39

Properties

No.	Parameter	Values	Units
1	Diameter outer (Casing inner)	0,1617	m
2	Diameter inner (Insulation outer)	0,080	m
3**	Heat flux*	266,5216 @10Yrs	W/m2
4**	Temperature in Horizontal section	5,77E+01 @10Yrs	C
5	Mass flow rate	8,8	kg/sec
6	Viscosity	0,0011	Pa.s
7	Density	998,554	kg/m3
8	Pr (lam,turb)	7,867/1,9	-

* This value was changed (1000x) in the simulation for practical reasons.

Data Source

HOCLOOP Deliverable D2.2 (Draft), accessed 06092023,

HOCLOOP Deliverable D2.1 (Draft), accessed 06092023,

W. M. Kays, 1994, ASME.

* Obtained from system simulation by VITO-IE-Team

vito.be

14 of 39

Results

III

Private & Confidential

15 of 39

f & Nu comparison (C=0%)

0,0231

0,0258

261,84

260,90

258,54

0,0282

Friction factor

Nusselt Number

vito.be

16 of 39

Eccentricity 0%

*dT is equivalent to changed Q value (1000x) used in the simulation for practical reasons.

vito.be

17 of 39

Eccentricity 10%

*dT is equivalent to changed Q value (1000x) used in the simulation for practical reasons.

vito.be

18 of 39

Eccentricity 20%

*dT is equivalent to changed Q value (1000x) used in the simulation for practical reasons.

vito.be

19 of 39

Eccentricity 30%

*dT is equivalent to changed Q value (1000x) used in the simulation for practical reasons.

vito.be

20 of 39

Eccentricity 40%

*dT is equivalent to changed Q value (1000x) used in the simulation for practical reasons.

vito.be

21 of 39

Eccentricity 50%

*dT is equivalent to changed Q value (1000x) used in the simulation for practical reasons.

vito.be

22 of 39

Eccentricity 60%

*dT is equivalent to changed Q value (1000x) used in the simulation for practical reasons.

vito.be

23 of 39

Eccentricity 70%

*dT is equivalent to changed Q value (1000x) used in the simulation for practical reasons.

vito.be

24 of 39

Eccentricity 80%

*dT is equivalent to changed Q value (1000x) used in the simulation for practical reasons.

vito.be

25 of 39

Eccentricity 90%

*dT is equivalent to changed Q value (1000x) used in the simulation for practical reasons.

vito.be

26 of 39

10% vs 80%

*T is equivalent to changed Q value (1000x) used in the simulation for practical reasons.

vito.be

27 of 39

*T is equivalent to changed Q value (1000x) used in the simulation for practical reasons.

10% vs 80%

vito.be

28 of 39

Variation of properties with eccentricity

*dT is equivalent to changed Q value (1000x) used in the simulation for practical reasons.

Friction factor

Nusselt Number

vito.be

29 of 39

Variation of Temp. with eccentricity

*T & dT is equivalent to changed Q value (1000x) used in the simulation for practical reasons.

vito.be

30 of 39

Can we increase further the Heat Transfer?

vito.be

31 of 39

Parametric study: bumps on the outer wall

Objective: to study affect on heat transfer and P losses

bump inwards (-)

baseline

bump outwards (+)

Bumps were generated using FFD Boxes.

vito.be

32 of 39

Alternative configurations: bumps on the outer wall

Same setup and BC of the baseline apply for the modified pipe geometries

Test case
-3	- 5,57
-2	- 3,71
-1	- 1,86
+1	+ 1,86
+2	+ 3,71
+3	+ 5,57

vito.be

33 of 39

Friction factor

decrease in f of

-5,48%

vito.be

34 of 39

Conclusions

Private & Confidential

35 of 39

Key take away points

CFD results co-relate well with VDI-Heat Atlas correlation and as per literature (for C=0%).

Nu-CFD calculations corelate well as per the relations in Ref. studies (for C=0%).

Letting the inner tube sit on the bottom of the casing (C=100%) will lead to ~75% decrease in heat transfer and ~25% decrease in pumping power (with Heat Flux BC).

Swirl generators to center the pipe.

vito.be

36 of 39

Way forward

Cold Fluid

Ground

insulation

Schematic Representation: VITO

Hot Fluid

vito.be

37 of 39

Thank you !

Dr.ir. Nitish Anand

nitish.anand@vito.be

Researcher

Energy Systems and Components Optimization (ESCO)

This project has received funding from the European Union’s Horizon Europe research and innovation program under grant agreement No. 101083558 (HOCLOOP).

hocloop.eu

vito.be

38 of 39

Thermo-hydraulic performance analysis of concentric pipes for geo-thermal application

N Anand, F Tosto, C de Servi, J van Bael.

Energy Systems and Components Optimization (ESCO),

Belgium.

This project has received funding from the European Union’s Horizon Europe research and innovation program under grant agreement No. 101083558 (HOCLOOP).

SU2 Conference 2024

vito.be

39 of 39

Grid Convergence Study

Mesh	dP (Pa)	%err	dT (K)	%err
1	4,56	0,7%	170,21	1,4%
2	4,53	0,2%	172,61	0,6%
3	4,52	-	173,71	-

selected grid

~1.5M elements

Temporal cost

10Hrs@48cores

vito.be