HomeFeaturesExamplesTutorialsHelpTheoryVerificationPricingContact
Theory manual

21. Thermal plume on the site plan

The thermal plume is the change of the ground temperature against the undisturbed ground, ΔT(x,y,t), in the plan around the borehole field. Authorities and neighbours ask how far the cooled (or warmed) ground reaches, whether it crosses the property line, and when it reaches a well, a neighbouring field or a building. The Thermal plume page answers these questions at the end of every month of the design period, from the monthly base loads of the design, with isotherms for chosen thresholds (default ±3 K) and distance checks against the objects of the site plan.

Two sources give the field: the analytic plume of this chapter, which runs after every calculation, and the temperature field of the 3D finite-element analysis when it is available. Both are sampled on the same raster and evaluated by the same procedure.

21.1 Load convention

qi [W/m] is the base ground load of month i per metre of borehole (> 0 heat injected): the monthly base load divided by the total length, or, with hourly loads, the mean of the hourly ground load over each 730-hour month. Month i starts at ti=i·730 h, and the map of month n is at its end:

ΔT(p,tn+1)=∑i≤n(qi−qi−1)Φ(p, tn+1−ti),q−1=0
(21.1)

where Φ is the response at the plan point p to a unit step of 1 W/m in the field. The peak loads are not included: they act for hours and do not reach the far field.

21.2 Analytic response: finite and moving finite line sources

Every borehole is a line source over its active length with heat rate wbq per metre. The receiver is the plan point averaged over the active depth range (from the shallowest top to the deepest bottom of the active lengths), or a short segment at a chosen depth. The response of an emitter at plan distance d is the segment-to-segment finite line source (Eskilson 1987, in the form of Claesson & Javed 2011), with receiver length Hr:

Φe(d,t)=F(d,t)2Hr·2πλ,F(d,t)=∫1/4αt∞1s2e−d2s2I(s)ds
(21.2)

where I(s) is the axial part, which holds the emitter and receiver depth ranges and the mirror image above the ground surface. With groundwater flow the moving finite line source (Molina-Giraldo et al. 2011) is used:

F(d,t)=e−2c(d−x)∫1/4αt∞1s2e−(ds−c/s)2I(s)ds,c=vT4α,vT=u(ρc)wC
(21.3)

with u the Darcy velocity, (ρc)w and C the volumetric heat capacities of the water and the saturated ground, and x the distance of p downstream of the emitter. The integral without the direction factor depends on d only; it is tabulated, and the factor e−2c(d−x)≤1 is applied per point, so neither part can overflow. The tables hold every month end on a logarithmic distance grid dj=rb,mine0.025j; between the nodes Φ is interpolated linearly in lnd (error about 10−4 relative).

21.3 Heat split between the boreholes

wb is the heat rate per metre of borehole b relative to the field mean, ∑bwbHb=∑bHb:

This is the long-time split. At early times the true split is more uniform (the boreholes do not interact yet); the difference concerns boreholes next to each other, not the far plume. Each borehole emits uniformly along its length. For a 3 × 2 field (6 m spacing, 100 m, 25 years) the corners take 1.07 and the middle boreholes 0.85 under a uniform wall temperature, 1.05 and 0.89 with Rb = 0.1 m K/W under a uniform fluid temperature.

21.4 Superposition

Per emitter type and distance node, Sj(n)=∑i(qi−qi−1)Φ(dj, tn+1−i) is a causal convolution over the months, done by the fast Fourier transform. Then, for every point p and month n,

ΔT(p,n)=∑swse−2c(dps−xps)[(1−ϕ)Sj(n)+ϕSj+1(n)],lndps=lnd0+0.025(j+ϕ)
(21.4)

one table lookup per point and borehole. A raster of 250 × 90 cells around a 4 × 4 field over 25 years (300 months) with groundwater takes a few hundredths of a second, so the plume is recomputed after every calculation and for every change of the thresholds, the depth or the objects.

Window. The raster covers the boreholes, the property-line polygons and the evaluation objects, plus a margin of max(10 m, e/4, 2.5αmaxtend) on every side (e the extent of these objects) and, with groundwater, twice that margin plus vT,maxtend more downstream, as about 150 × 150 square cells. A region that touches the edge of the raster is flagged, and the window is enlarged by half its size on the sides concerned (up to three times) and sampled again, so the extents are not cut by the window.

21.5 Distance checks

The field is sampled at the cell centres and at points every 1 m or less along the objects. Then, for each threshold θ (cooling θ<0: the region ΔT≤θ; warming θ>0: ΔT≥θ):

The 3D source evaluates the finite-element temperature field of the 3D analysis at the same plan points, at 12 depths over the active depth range (averaged) or at the chosen depth, with the same load convention. The 2D and 3D maps of a month can therefore be compared and overlaid; outside the finite-element block there is no value.

21.6 Verification

The analytic plume was checked against an independent quadrature of the point-source integral along every borehole (finite and moving finite line source with the image above the surface), sampled on the same raster and evaluated by the same code, and against the 3D plume of the same project. Point values: within 0.18 % at a single depth and 0.09 % averaged over the depth for a single borehole over 25 years, within 0.5 % for seasonal loads and with groundwater, and +0.05 to +0.17 % for an inclined borehole. Table 33 shows isotherm extents for a constant extraction of 20 W/m (ground 2.5 W/(m K), 100 m boreholes, 25 years, at 54 m depth for the reference columns).

Table 33. Isotherm extents of the analytic plume against the independent reference and the 3D plume.
CaseIsothermQuantityReferenceAnalytic plume3D plume
single borehole−2 Kreach [m]8.578.578.45
single borehole−2 Karea [m²]230.6230.6223.1
single borehole−0.5 Kreach [m]32.7932.7932.58
single borehole−0.5 Karea [m²]337733773304
3 × 2 field, 6 m−3 Kreach [m]29.6629.6629.21
3 × 2 field, 6 m−3 Karea [m²]341434153356
3 × 2 field, 6 m−0.5 Kreach [m]61.9861.9963.24
3 × 2 field, 6 m−0.5 Karea [m²]133691337213575

The analytic plume reproduces the reference to 0.01 m in every extent and to 0.1 % in the areas; the one exception is a region of 5 m² on cells of 1.1 m (strong groundwater flow), where the raster resolution gives −3 % area. The 3D plume lies within −4.3 to +2.0 % in reach and downstream length and −6.6 to +1.5 % in area of the analytic one: slightly smaller at the near isotherms and with groundwater, slightly larger at the far isotherms. No isotherm touched the edge of the raster in any case.

21.7 Limits