This study addresses the rock mechanics and construction problems inherent in constructing a series of annular tunnels in a geometric array following the inner surface of a toroid at a depth of about 300 metres in dolomitic limestone. Rock behavior under cyclic loading, tunnel support, lining, drainage, tunnel separation, construction methods and cost are examined.
Studies have been conducted over a period of several years by a project team at the University of Wisconsin - Madison to develop the concept of electromagnetic energy storage in a cryogenically cooled superconducting solenoid. Because of the high containment loading generated by the magnetic field, the most economic containment is an underground structure where the rock mass can be used to sustain the loads. Until recently, the major effort has been devoted to electromagnetic and cryogenic research, but, as reported by Haimsen, Doe and Fuh at Rockstore 77, attention is now being given to a more detailed examination of the civil engineering work involved. The initial concept involved an arrangement of high narrow annular tunnels excavated by drill and blast methods in granite. The analysis of the loading imposed by the energized superconductors indicated that high shear loads would have to be supported by the tunnel walls and that zones of tensile stress would call for extensive rock reinforcement to prevent rock failure. Moreover, the excavation of these tunnels by conventional drill and blast methods would necessarily result in a high degree of uncertainty as to the elastic behavior of the rock mass under the cyclic loading applied by the operation of the electromagnet. After investigation of alternative solutions to avoid some or all of these problems, it was decided to examine the possibility of using multiple machine-bored tunnels in a configuration following the current sheet surface. The medium chosen for the machine-bored tunnel study was the Sinnipee dolomite which is present at the right depth and thickness on the western shore of Lake Michigan north of Milwaukee. The properties of this material were known from other investigations in sufficient detail for this preliminary study and in particular, ample successful experience is available regarding the construction of machine-bored tunnels in carbonate rocks at rates of advance likely to lead to economical construction. The typical tunnel array is shown in Fig. 1 below with numerical values for the leading dimensions tabulated in Table 1 for various possible configurations. While there appears to be a substantial reduction in unit cost of construction per KWh of capacity as the total capacity and magnetic field strength are increased, it was determined that the initial work would be limited to a 1000 MWh facility with a field strength of 2.5 Tesla as being the most likely size to be built first.
One of the most troublesome aspects of the work previously reported (Haimson, Doe & Fuh, 1977) was the amount of strain assumed in conducting the finite element analyses for the original configuration of the containing structure in conventionally excavated granite.
