@

2. Review of Excavation Tests

Over the past four decades, extensive excavation tests by disc cutter have been performed. In several of these studies, thrust force and rolling force was measured for various rock types, spacing of cutter and cutting depth.2.1 Effects of rock strength

In several studies, compressive strength was considered as an index of rock type. Thrust force was found to be proportional to compressive strength (Roxborough and Phillips, 1975; Nishimatsu et al., 1975; Snowdon and Ryley, 1982). In a few studies, total hardness (Nelson et al., 1983; Tarkoy, 1983), fracture toughness (Fangming et al., 1992) or critical energy release rate (Nelson and Fong, 1986) is used as an index of rock type. In addition, thrust force was found to be proportional to these properties.

2.2 Effects of cutting depth

Thrust force increases as cutting depth increases. Previous experimental results are divided into two categories, as follows

a) Thrust force is proportional to the square root of cutting depth

Laboratory test: Takaoka (1968), Ozdemir (1984), Sanio (1985), Fangming (1992)

In-situ test: Samuel (1984)

b) Thrust force is proportional to cutting depth.

Laboratory test: Roxborough (1973), Nishimatsu (1975), Howarth (1982), Snowdon (1982)

In-situ test: Gaye (1972)

Variations in the experimental conditions, such as spacing of cutter and cutting depth, may explain these contradictory findings.

2.3 Relationship between rolling force and thrust force

Several studies have reported that the ratio of rolling force to thrust force is proportional to the square root of cutting depth and is independent of rock properties (Takaoka, 1968; Sanio, 1985).

2.4 Equation for estimating rock strength

The findings for excavation tests reported in previous studies can be summarized as follows:

a) Thrust force F1 is proportional to the compressive strength of rock Q

b) Thrust force is proportional to either cutting depth p or its square root.

c) The ratio of rolling force F2 to thrust force is proportional to the square root of cutting depth.

The relationship between thrust force and cutting depth remains unclear. However, in the present study, thrust force is assumed to be proportional to cutting depth. Therefore, F1 and F2 can be written as

F1 = a1 Q (1)

F2 = a2 Q p^1.5 (2)

where a1 and a2 are constants and Q is rock strength, which is dependent on faults or cracks. The thrust of TBM FN is the sum of thrust forces F1 and the torque of TBM TR is the sum of rolling forces multiplied by the radius of rolling. Therefore, FN and TR can be written as

FN = c1 Q p (3)

TR = c2 Q p^1.5 (4)

where c1 and c2 are constants. In Equations (3) and (4), Q is the average rock strength of a face. From Equations (3) and (4), Q can be written as

Q = FN/(c1 p) (5)

Q = TR/(c2 p^1.5) (6)

thus, rock strength can be estimated using Equations (5) and (6) if c1 and c2 can be determined. Thrust and torque are usually measured in order to prevent TBM from exceeding the design load. Cutting depth can be calculated from penetration rate and cutter head rotation, which are always measured. Rock strength can be measured using the measuring system normally used for TBM.

In Equation (5), Q is proportional to the field penetration index (FPI), introduced by Nelson (1983). Although several studies have examined the relationship between FPI and the properties of rock, a real-time monitoring system, such as that using Equation (5), has never been proposed.