is Íslenska en English

Lokaverkefni (Meistara)

Háskóli Íslands > Verkfræði- og náttúruvísindasvið > Meistaraprófsritgerðir - Verkfræði- og náttúruvísindasvið >

Vinsamlegast notið þetta auðkenni þegar þið vitnið til verksins eða tengið í það: http://hdl.handle.net/1946/22923

Titill: 
  • Titill er á ensku Experimental investigation of induced seismicity in granitic rock on centimeter scale
Námsstig: 
  • Meistara
Útdráttur: 
  • Útdráttur er á ensku

    Enhanced Geothermal Systems (EGS) permit the exploitation of geothermal energy in deep, dry and impermeable formations of the Earth’s crust. To create permeability, fluids under high pressure are injected into these deep formations, causing injection-induced seismicity. If large-magnitude seismic events occur, they can lead to public nuisance and infrastructure damage. The desire to better understand the physical processes causing large-magnitude events, so that they can be avoided or mitigated, is the motivation of this thesis.
    The main focus here is on the localization and the associated analysis of differential stress- and injection-induced seismic events on a laboratory scale. Granite samples were subjected to axial compression under confinement until shear failure occurred. Detected differential-stress induced small-scale seismic events, referred to as acoustic emissions (AE), were localized; these made microcrack initiation, fault nucleation, and fault propagation visible. Magnitude analysis of induced AE showed temporal b-value variation throughout the experiment. Also, AE were detected during fluid injection into uniaxially loaded, cylindrical granite samples until breakdown pressure was reached. Breakdown pressure was reached several times over multiple pressure cycles with the result that injectivity (i.e. pressure vs. injection rate) and radial strain increased with every repeated cycle. This provided evidence of irreversible de-formation. Estimated b-values obtained over pressure cycles appeared to be of a similar range of magnitude. Temporal b-values over one pressure cycle show a decrease with increasing injection pressure and are lowest at breakdown pressure. Accuracy of localized induced AE was comparably low and the main reason for this turns out to be the poorly developed seismic velocity model required for localization: it does not adequately describe occurring heterogene-ities and anisotropies due to microcracking or fluid injection.

Samþykkt: 
  • 17.9.2015
URI: 
  • http://hdl.handle.net/1946/22923


Skrár
Skráarnafn Stærð AðgangurLýsingSkráartegund 
Thesis_Villiger.pdf6.24 MBOpinnHeildartextiPDFSkoða/Opna