Daniel Swenson

Fluid circulation and heat extraction from engineered geothermal reservoirs

Abstract A large amount of fluid circulation and heat extraction (i.e., thermal power production) research and testing has been conducted on engineered geothermal reservoirs in the past 15 years. In confined reservoirs, which best represent the original Hot Dry Rock concept, the flow distribution at any given time is primarily determined by three parameters: (1) the nature of the interconnected network of pressure-stimulated joints and open fractures within the flow-accessible reservoir region, (2) the mean pressure in the reservoir, and (3) the cumulative amount of fluid circulation—and therefore reservoir cooling—that has occurred. For an initial reservoir rock temperature distribution and mean fluid outlet temperature, the rate of heat extraction (i.e., thermal power) is at first only a function of the production flow rate, since the production temperature can be expected to remain essentially constant for some time (months, or even years). However, as reservoir circulation proceeds, the production temperature will eventually start to decline, as determined by the mean effective joint spacing and the total flow-accessible (i.e., heat-transfer) volume of the reservoir. The rate of heat extraction, which depends on the production flow rate, can also vary with time as a result of continuing changes in the flow distribution arising from reservoir cooling. The thermal power of engineered reservoirs can most readily be increased by increasing the production flow rate, as long as this does not lead to premature cooldown, the development of short-circuit flow paths, or excessive water losses. Generally, an increase in flow rate can be accomplished by increasing the injection pressure within limits. This strategy increases the driving pressure drop across the reservoir and the mean reservoir pressure, which in turn reduces the reservoir flow impedance by increasing the amount of joint dilation. However, the usefulness of this strategy is limited to reservoir operating pressures below the fracture extension pressure, and may lead to excessive water losses, particularly in less-confined reservoirs. Under such conditions, a downhole production-well pump may be employed to increase productivity by recovering more of the injected fluid at lower mean reservoir operating pressures.

Finite element analysis of edge cracking in plates

We present a finite element analysis of cracking due to edge loading of brittle plates. In the analysis, the crack path is not specified a priori, but is predicted as an integral part of the analysis. Automatic remeshing is performed to follow the moving crack. Comparison with experimental data of cracking in PMMA demonstrates the capability of the analysis to predict curving crack propagation. This verifies that a maximum circumferential stress criterion (equivalent to a zero Mode II condition) can be used to predict crack propagation direction.

The effects of thermal deformation on flow in a jointed geothermal reservoir

Abstract In a Hot Dry Rock (HDR) geothermal reservoir, heat is removed primarily by fluid flowing through joints in relatively impermeable rock. Cooling causes the rock to thermally contract and deform, changing the joint opening; fluid viscosity can change by an order of magnitude over typical temperature ranges. In this paper we examine these effects on flow in both a single joint and in a more complex reservoir. We show that thermal deformation can either reduce flow due to local pinching of a joint or can increase flow due to global thermal opening of a joint. Which effect will dominate is dependent on the peripheral boundary conditions. In a reservoir, it is likely that both behaviors occur. Data from the HDR site at Fenton Hill indicates a generally decreasing flow (pinching). However, there have been periodic step increases in flow that more than compensate for the reduced flow (global joint opening).

Effect of Thermal Deformation on Fracture Permeability in Stressed Rock Masses

Abstract We analyse the effect of thermal contraction of rock on fracture permeability. The analysis is carried out by using a 2D FEM code which can treat the coupled problem of fluid flow in fractures, elastic and thermal deformation of rock and heat transfer. In the analysis, we assume high-temperature rock with a uniformly-distributed fracture network. The rock is subjected to in-situ confining stresses. Under the conditions, low-temperature fluid is injected into the fracture network. Our results show that even under confining environment, the considerable increase in fracture permeability appears due to thermal deformation of rock, which is caused by the difference in temperature of rock and injected fluid. However, for the increase of fracture permeability, the temperature difference is necessary to be larger than a critical value, ΔTc, which is given as a function of in-situ stresses, pore pressure and elastic properties of rock.

New Wayfinding Techniques in Pathfinder and Supporting Research

This paper presents the wayfinding and door selection algorithm used by the Pathfinder egress simulator. The development of Pathfinder’s Locally Quickest algorithm will be discussed along with unforeseen consequences and one promising, but discarded, initial attempt. Validation of this new model is presented as well as characteristics of the third party research that make simulator validation possible.

Polymer reinforcements for retarding fatigue crack growth in metals

Pure rolled, annealed copper and copper-polyimide (Kapton) laminates were tested under constant amplitude cyclic loading to determine the fatigue crack growth rate. It was found that the laminated samples could sustain a much higher load for the same fatigue life, or had a longer fatigue life for the same load. This result is due to the polyimide bridging across cracks in the copper. The reduction of the crack tip stresses due to bridging is quantified by an analytical approximation and by layered finite element analyses. Despite the low elastic modulus of polyimide relative to copper, the stress reduction is significant due to the high effective stiffness of the bridging layer which results from the thinness of the adhesive layer between the copper and polyimide. When the experimental data from the laminated samples are analyzed using the results of a layered finite element analysis good correlation is obtained between the crack growth rate in pure copper and in the copper-polyimide laminates.