Experimental study of blasting excavation for large cross-section 1 tunnel in horizontal layered rock mass 2

The drilling and blasting method is still the main method in mountain tunnel excavation. For large cross- 8 section tunnel in horizontal layered rock mass, tunnel blasting often causes serious overbreak and underbreak. In this 9 study, blasting excavation tests of tunnel upper face were conducted and failure mechanisms of surrounding rocks with 10 weak beddings and joints were analyzed based on the Panlongshan tunnel. Then, the blasthole pattern, the cut mode, a 11 variety of peripheral holes, the charge structure and the maximum single-hole charge were optimized. Compared with 12 the failure characteristics, overbreak and underbreak, and deformations of surrounding rocks before and after 13 optimization, the latter was better in tunnel contour forming and surrounding rock stability. The results show that after 14 optimization, the large-area separation of vault rock mass is solved, the step-like overbreak of spandrel rock mass is 15 reduced and the large-size rock blocks and underbreak are avoided. The maximum linear overbreak of vault, spandrel, 16 and haunch surrounding rocks is decreased by 42.3%, 53.7% and 45.1%, respectively. The underbreak at the bottom of 17 the upper face is reduced from -111.5 to - 16.5 cm. The average overbreak area is decreased by 61.1%. In addition, the 18 displacements after optimization finally converge to the smaller values. The arch crown settlement and the horizontal 19 convergence of haunch are reduced by about 21.6% and 18.3%, respectively. Furthermore, from the completion of 20 blasting excavation to the stabilization of surrounding rock, it takes less time by using the optimized blasting scheme.

timing sequence control techniques. Taking into account the specified control indices, including the arch crown settlement, thickness of the blasting damage zone, Liu and Liu (2017) put forward an intelligent optimization 49 method of smooth blasting parameters for mountain tunnels by using a GA and ISVR coupling algorithm. Salum 50 and Murthy (2019) suggested overbreak control methods by means of optimizing blasthole distance and the 51 thickness of smooth blasting rock layer, as well as the charge of peripheral hole.

52
Tunnel excavation in the horizontal layered rock mass is highly influenced by the complex weak structural 53 planes (Solak 2009;Deng et al. 2014). Some related studies indicate that under this geological structure, the 54 explosive stress wave will be blocked from the structural planes, causing damage to the remaining rock mass out

62
Although there are many studies focused on the overbreak control for tunnels in jointed rock mass by proper 63 blasting parameters design. Unfortunately, as a result of the complexity of influencing factors of tunnel blasting 64 excavation and lack of understanding of the weak structural planes, there are no specific criterions and methods to 4 1.31 is used in the transverse direction, whereas in the longitudinal direction, the concentrated charge at the bottom 120 of blasthole is adopted. each part is clustered and detonated by detonators. The blastholes in the haunch is divided into left and right parts, 125 and each part is clustered and detonated by detonators.

126
To sum up, the parameters of blastholes and charges of upper face are listed in Table 2.

127
[ From ZK80+263.0 to ZK80+239.6, the left tunnel was excavated 6 times by using the above blasting scheme.

131
The failure characteristics of surrounding rock after blasting excavation are shown in Fig. 6. It can be seen from 132 the arch remaining rock mass ( Fig.6a, b, c), discontinuous horizontal beddings and vertical joints are developed.

133
The bedding spacing ranges from a few centimeters to tens of centimeters. These joints are intersected with 134 horizontal beddings, and the joints are filled with mud. Affected by the influence of horizontal beddings and joints, 135 the surrounding rock was badly damaged and the tunnel contour was very irregular. The vault rock mass fell off 136 along a bedding plane, forming a flat outline. At the spandrel, broken rocks slid down along a joint plane, leading 137 to a distinct step-like outline. The maximum height and width of step-like rock fracture surface were 72 and 34 138 cm.

139
It could be seen from the bottom remaining rock mass (Fig.6d, e), the face was roughness and there was 140 underbreak at the bottom. Because the distance of the one-wedge cut holes on the face was about 8.0 m, the cutting 141 blasting generated large-size stone. The length, width, and height of the stone was about 1.8 m × 1.0 m × 1.4 m.

142
In order to realize the transportation of large-size stone, it was necessary to carry out secondary drilling and The sum of the stress waves entering rock at interfaces A and B can be expressed as:

263
The detonating cord goes through the entire length of the holes. The opening of the holes was blocked with a length 264 of 250 mm stemming.

265
Finally, the optimized parameters of blastholes and charges of upper face are given in Table 3.

285
-16.5 cm. The advantage was that the bottom was flat, which was conducive to the next tunnelling.

286
The average length of each round excavation was 3.76 m. An average blasting efficiency was 94.0%, which 287 was increased by 2.7% through a comparison with the previous result.

288
[ Figure 17 near here]    Table 3 The optimized blasthole and charge parameters of the upper face in the Class-IV surrounding rock 527