In-situ measure to internal stress of shotcrete layer in soft-rock roadway

In order to solve the difficult conditions of soft rock, water-trickling and hard-maintain of main air-return roadway in Tarangaole Colliery, high pretensioned stress and intensive bolt-shotcrete support program was designed and mechanical property of shotcrete layer was specially monitored through utilizing a type of concrete stress meter with oscillating chord after the program was carried out. It was indicated that, due to rock pressure and support resistance, the interior of shotcrete layer would emerge diverse stresses in axial, radial and tangential directions. With time passing internal stresses in three directions, whose average values were −0.061, 0.043 and 0.517 MPa respectively, fluctuated first and then tended to stability slowly. The axial and radial stresses were relatively smaller than tangential stress which was 11, 12 times the two formers respectively. Along the section of roadway, axial and tangential stresses distributed symmetrically and increased gradually from the top of arch to the waist of wall, but reduced at the foot of wall. Radial stresses reduced from the top of arch to the waist of arch first, and then increased in the waist of wall. Axial stresses were tensile substantially, except for stresses in arch vault tending to compressive, but all the radial stresses were compressive. Nevertheless, tangential stresses in the wall were compressive and tangential stresses in the arch were tensile. During the period of roadway excavating, the stress of shotcrete layer was less than its ultimate bearing capacity, with no significant stress concentration. At the end of this article, some suggests are given to shotcrete support design.


Introduction
Bolt-shotcrete is considered as an effective and economical support to colliery roadway (He and Sun 2004;Jing and Li 2008). After the roadway is supported by bolt-shotcrete, it is very important to track and monitor the working status of bolt-shotcrete structure, especially the sprayed concrete layer (shotcrete layer), whose stress and deformation conditions are directly related to roadway safety. However, there are few literatures about mechanical site-measure to shotcrete layer in colliery roadway, but a few ones in highway tunnel, railway tunnel, side slope, hydraulic tunnel, dam foundation and underground engineering (Lai et al. 2005;Jia et al. 2006;Li et al. 2007). Their measure tools and methods are more diverse and field data's are more useful, which can be a reference to shotcrete layer measure.
Soft rock, water-trickling and hard-maintain exist in main air-return roadway of Tarangaole Colliery in Erdos City. In order to overcome these difficulties, high pretensioned stress and intensive bolt-shotcrete support program was designed and mechanical property of shotcrete layer was specially monitored through utilizing concrete stress meter with a type of oscillating chord after the program was carried out in the roadway (Kang 2000;Li et al. 2007;; . The measurement items included internal axial stress, radial stress and tangential stress in shotcrete layer. Measure results showed that, during the period of roadway excavating, the stress of shotcrete layer was less than its ultimate bearing capacity and its deformation was also limited, those two aspects of which indicated that the high pretension press and intensive bolt-shotcrete support technology met the surrounding rock control requirements for main air-return roadway in Tarangaole Colliery.
2 Project overview

Production and geology conditions
Tarangaole Colliery is an constructing mine, designed shaft developing and plate extracting, whose industrial square sets out the main shaft, auxiliary shaft and central air-return shaft. Currently three shafts were fully completed and two main air-return roadways located in the east and west sides of air-return shaft are being tunnelled. This measure site is just selected in the part of east air-return main-roadway, whose layout is shown in Fig. 1.
East air-return main roadway is about 570 m depth, excavated along the roof of coal seam 3-1. Coal seam 3-1 is served as the main seam in Tarangaole Colliery, whose dip is 1°-5°and thickness is 0.21-6.50 m (average 2.91 m). The roof lithology of coal seam 3-1 is mainly sandy mudstone or siltstone, locally fine-grained sandstone or coarse sandstone and the floor lithology is mainly sandy mudstone. Average compressive strength of rock is 14.5 MPa, moisture content is 0.81 %-13.65 % and softening coefficient is 0.05-0.51. Specific lithologies of roof and floor are shown in Fig. 2.

Support program
The shape of east air-return main-roadway is designed to semicircle, wall height 1,800 mm, arch height 2,850 mm, bottom width 5,700 mm and basal area 23.0 m 2 . According to the production and geology conditions, high-pretensionpress & intensive & once Support theory and the   (Kang 2000;GB 50086-2001;Kang et al. 2008Kang et al. , 2009, High Pretensioned Stress and Intensive Bolt-Shotcrete Support program is adopted. Specific parameters are as follows:

Measure design
Due to rock pressure, the interior of shotcrete layer will emerge diverse stresses along the axis of roadway, perpendicular to roadway axis (i.e. roadway radial, Zheng 1988) and the peripheral tangential of roadway. In order to study the actual stress properties of shotcrete layer, a method of installing integrated measure stations is designed to monitor the three internal stresses (Wang 1988;Xu 2009;Li et al. 2010).  Figure 4 shows measure points arrangement of one section.

Measure content, instrument and method
The contents of this measure test are shotcrete layer internal axial stress, radial stress and tangential stress, including stresses variation with time and spatial distribution patterns. GSJ-2A concrete stress meter with oscillating chord is utilized as measure instrument. This type of stress meter comprises a pressure disc and a vertical chord sensor, associated with a frequency detector. Its working principle is: the disc produces stress in effect of vertical press; stress is converted to a frequency signal by the sensor; frequency signal is measured by the detector. The stress can be calculated through the following formula: where P is the internal stress of shotcrete layer, MPa; f and f 0 are measure frequency and initial frequency of the sensor respectively, Hz; A and B stand for calibration coefficient, MPa/Hz 2 or MPa/Hz 2 . The methods and steps of this measure test are shown in Fig. 5. Firstly, the pressure disc of stress meter should be fixed to steel mesh along different directions with the wire. Then concrete is sprayed and the meter must be completely buried in concrete. Afterwards, measure wires should be lead to the unaffected area along the roadway surface and the connectors should be wrapped with waterproof canvas. Finally, through the frequency detector, dates of shotcrete layer stress are measured.

Measure result and analysis
After measure station is installed, it takes nearly three months to monitor the stress dates which are listed in Table 1. Figures 6 and 7 show stress measure values in time and space respectively. 4.1 Shotcrete layer stress variation with time Figure 6 is temporal variation curve of internal stresses of shotcrete layer. As can be seen from Table 1 and Fig. 6: (1) Throughout the measure period, internal axial stress of shotcrete layer fluctuated firstly and then tended to   In-situ measure to internal stress of shotcrete layer 325 stability whose average value was -0.059 MPa (tensile stress). In the initial half month, in addition to the vault and waist of arch, stresses of remaining five measure The average value of axial stresses was -0.061 MPa.
(2) As the same with axial stress, internal radial stress of shotcrete layer stabilized slowly after fluctuation. Unlike the former, the latter changed smaller (0.043 MPa) and substantially compressive stress.
Only the stresses of left wall back, left wall foot and right arch waist skirting were tensile early, but latter converted to compressive. In the initial month, remaining stresses fluctuated in growth and the maximum was 0. (3) Each internal tangential stress of shotcrete layer varies. These stresses of four measure points, which were located in the lower part of roadway, quickly increased to 0.6 MPa or more within 5 days. Tangential stress of right wall foot reached a maximum (1.425 MPa) 0.5 days later, these stresses remained changing stably, by less than 5 %. The remaining stresses of three measure points, which were located in the higher part of roadway, did not change significantly and always maintained slight fluctuations (no more than 0. 4.2 Shotcrete layer stress spatial distribution Figure 7 shows final stress spatial distribution along the roadway surrounding in 78th day. As can be seen: (1) Overall, though the stresses in three directions were not consistent with time variation, their spatial distribution had largely followed a common feature, that is all of the three internal stresses distributed evenly and symmetrically along the section of roadway, with no significant stress concentration. Axial and tangential stresses increased gradually from the top of arch to the waist of wall, but reduced in the foot of wall. While, due to weight of shotcrete, radial stresses reduced from the top of arch to the waist of arch firstly, and then increased in the waist of wall. It is thus clear that force and displacement of the arch-wall connection are both larger than the other part, which should be shotcreted to be thicker.
(2) The axial and radial stresses were relatively smaller than the tangential stress. The latter was 11 times, 12 times the two formers respectively. However, considering the characteristics of the measure method, the tangential stress measured here actually included some weight of shotcrete layer. If the impact of weight was excluded, the gap between the three stresses would decrease.
(3) Axial stresses were substantially tensile stresses, except for stresses in arch vault tended to compressive. Nevertheless, all of radial stresses were compressive stresses. Tangential stresses in the wall were compressive, but tangential stresses in the arch were tensile. This is the reason that cracks and damages often appear at the top of roadway sprayed by concrete. Therefore tangential tensile strength of top shotcrete should be designed to enhance.

Conclusions
(1) Due to rock pressure and support resistance, the interior of shotcrete layer will emerge diverse stresses in axial, radial and tangential directions within short time just after the concrete was sprayed, which can be monitored through utilizing a type of concrete stress meter with oscillating chord. Shotcrete can improve stress state of surrounding rocks and the roadway should be shotcreted timely after it was excavated.
(2) With time passing, internal stresses in three directions fluctuated firstly and then tended to stability, whose average values were -0.061, 0.043 and 0.517 MPa respectively. The axial and radial stresses were relatively smaller than tangential stress, which was 11 times, 12 times the two formers. (3) Axial and tangential stresses increased gradually from the top of arch to the waist of wall, but reduced at the foot of wall. While, due to weight of shotcrete, radial stresses reduced from the top of arch to the waist of arch firstly, and then increased in the waist of wall. All of the three internal stresses distributed evenly and symmetrically along the section of roadway, with no significant stress concentration. (4) Axial stresses were tensile substantially, except for stresses in arch vault tended to compressive, but all of radial stresses were compressive. Nevertheless, tangential stresses in the wall were compressive and tangential stresses in the arch were tensile. (5) Stress and deformation of the arch and wall connection are both larger than other parts, and tangential stress of arch top is tensile. The two parts should be shotcreted to be thicker and designed to be stronger.
Open Access This article is distributed under the terms of the Creative Commons Attribution License which permits any use, distribution, and reproduction in any medium, provided the original author(s) and the source are credited.