| Abstract |
On December 13th, 2017, the first subhorizontal geothermal well, a concept first published in February, 2011 at the Stanford geothermal reservoir engineering workshop, has been completed on the Paris suburban Cachan site. The well architecture, further duplicated on March 2018 on the injector well, pioneered a world premiere in geothermal engineering, as it achieved two 1 001/1 005 m long, 87 to 93° slanted openhole drains in a Mid Jurassic target carbonate reservoir at a 1 550 m true vertical depth. Project workflow is analysed from the conceptual modelling/well design adequacy; drilling/well placement, wireline logging and geochemical monitoring inputs, well stimulation/testing, final doublet performance and reservoir assessment stand points. Prior to drilling a relevant conceptual model integrating available geologic and hydrodynamic reservoir features was required. The multi layered reservoir structure is hosted by a carbonate platform of Mid Jurassic age displaying complex facies distribution related to the depositional environment, microfracturing and diagenetic processes. Hence correlation of nearby offset wells enabled to derive a so-called multi-layered equivalent sandwich model including all pertinent geometric, hydrothermal and geochemical attributes in which candidate well trajectories could fit in. Both wells are quasi identical in design by sharing an upper, dual cased (16"x10"3/4), 18"1/2 vertical phase followed by a deviated section initiated by a 14"3/4 in an arc trajectory, under a standard MWD-PDM assembly, and finalised by a ca 1 000 m long 8"1/2 subhorizontal drain drilled under a geosteered LWD-MWD-RSS (Rotary Steerable System) BHA, securing constant rotation (no sliding) and upgraded hole calibration. The geosteering strategy addressed a two step sequence. (i) While drilling, real time data acquisition. • Directional drilling= monitoring/controlling RSS downhole tool performance. • LWD tool string= Gamma Ray, Neutron porosity, multi-frequency resistivity, imaged azimuthal density. • XRD, XRF mineralogic and elemental analysis. • Mud Logging cutting petrography. (ii) Post drilling analysis. • Integration of Wireline NMR and Sonic Dipole. • Production logging (PLT) and flow metering tools. Regarding the wireline input, tool tractor drive limitations could be (partly) compensated thanks to LWD tracking back up, which supplied useful clues while steering drain trajectories, in particular on well n°2 characterised by thin, metric size, up dip varying, bed layers. On well n°1 successful PLT spinner flowmetering provided unvaluable insight on flow and temperature profiles along the entire drain, a key information in calibrating a wellbore heat transfer model. Noteworthy was the first application of Nuclear Magnetic Resonnance (NMR/CMR tool) and dipole sonic (DIS tool) which were most effective in (i) correlating porosities to permeabilities (and VV) along with assessing lateral extents of thin bed layering from P and S wave sources, and (ii) matching drain porous segments eligible to stimulation via coiled tubing acid spotting. XRD/XRF real time geochemical monitoring proved rewarding with respect to (i) candidate alkaline (Sr, Na, Mg) and mineral (Mn, Fe, Zn) proxies as porosity and diagenetic markers, and (ii) metal oxide lithofacies indicators. Surprisingly, well testing, in spite of locally complex when not adverse well and reservoir attributes among which (i) a stratified structure intercepted by a subhorizontal, occasionally tortuous, drain profile, (ii) a non homogeneous flow distribution along the drain, (iii) interlayer cross flow, and (iv) interfered pressures and temperatures induced by neighbouring GDH doublets operating at (winter) seasonally maximum flow ratings, matched satisfactorily the idealised horizontal well test theory, particularily the pressure derivative within the early radial and pseudo-radial drainage regimes. As a result it made it possible to calibrate and reliably operate wellbore heat flow, well test and reservoir simulation models. Summing up, the Cachan project achieved technical and economic viability of the subhorizontal well concept in a multi-layered carbonate reservoir setting and densely populated and drilled environments. By demonstrating a 450 (nominal)/500 m3/h (maximum) rated capacity it is due to replace two existing, 34 yr aged, doublets cumulating a maximum 350 m3/h rating. It elsewhere proved a thorough reservoir evaluation tool. The extension of the subhorizontal and radial well architectures and related RSS and sharp angle drilling/geosteering technologies to similar low permeability stratified and shallow fast cooling breakthough settings is discussed in fine. |