Introduction

Coal is located in a certain position of underground rock mass, contacts with the surrounding rock mass, and maintains its stress equilibrium state. After the underground coal seam is extracted, the strata around the mining space lose their support and gradually move to the panel and destroy. With the continuous expansion of mining space, this moving process gradually expands from the position of the mining space to the periphery and upward, until it spreads to the surface, causing surface subsidence and horizontal movement. With the acceleration of urbanization, a large number of coal mines in China engage in mining under villages. The panel formed by underground coal mining is transmitted in the ground via the rock strata, thereby causing damage to the buildings and structures on the surface.

Building movement and deformation are monitored using external methods [terrestrial laser scanning (TLS), total station, and differential level] to master the mechanical state of internal structures and walls in ground deformation areas. On the basis of deformation, a numerical simulation software is used to analyze the internal mechanical state of buildings. Understanding the internal stress state of an intact or slightly damaged building can guide its reinforcement and prolong its life. It can also support the design of anti-deformation buildings through simulated reference data and provide a basis for the design of anti-deformation houses.

In recent years, TLS and other auxiliary methods have been utilized to detect deformation and damage in objects such as bridges, dams, buildings, and landslides (Table 1). The traditional differential leveling, close-range photogrammetric survey, and interferometric synthetic aperture radar (InSAR) technique are also used to monitor ground subsidence, and some applications are shown in Table 1.

Table 1 Some monitoring methods on ground disaster and subsidence
Full size table

Numerical simulation has been utilized in many areas of mining research (Table 2). Mining methods include long-wall mining, partial mining, and so on. Software for numerical simulation includes commercial types (e.g., FLAC3D, ANSYS, ABAQUS, COSFLOW, and UDEC) and self-developed types (e.g., discrete element and finite-element method). Numerical simulation of long-wall mining is performed to study the evolution of rock strata and surface displacement, whereas that of partial mining is conducted to study the stability of coal pillar and the deformation law of strata and surface.

Table 2 Studies that performed numerical simulation of underground mining
Full size table

Underground mining extraction causes deformation of the rock mass and the surface, which may cause damage to the construction of house foundation and other facilities. To consider the interaction between the manhole and the horizontally strained soil, Zięba and Kalisz (2019) studied the impact of horizontal soil strain on flexible manhole riser deflection based on laboratory test results. Use of a sliding joint is an effective method to decrease the stress in foundation structure where there is a horizontal deformation of subsoil or horizontal deformation of a foundation structure, Cajka et al. (2011) found that temperature is dependent with shear resistance, and the sliding joint shear resistance as a function of deformation velocity and temperature, which is used for the numerical analysis of stress/strain relation between foundation structure and subsoil. Deck et al. (2003) described soil–structure interaction phenomena at work during a mining subsidence. A finite-element software is used to generate models incorporating the ground material and a supported rigid structure. The structures are loaded in particular by ground curvature and strain. The cases where ground movements are integrally imparted to the structure are distinguished from those showing a strong interaction. The problem is simplified by dissociating the ground curvature and the strain. Chang et al. (2011) used the finite-element models ANSYS and given the soil–structure interaction phenomena produced at the interface between the reinforced concrete structure and the soil have been used to perform a parametric study of the strengthening design of the foundation and the sluice chamber within mining subsidence area.

In the underground coal-mining area, the external surface of the disturbed building and the ground monitoring are studied separately, and a lot of numerical simulation is used to study the stress and displacement of the building. There is a lack of a combination of them. Therefore, the influence of the ground movements of the coal-mining area on the internal mechanics and the displacement state of the building is the focus of this paper.

External monitoring of houses in subsidence area

Geological and mining condition

The study area is located in Huainan, north-central Anhui Province, China. Huainan is a coal enrichment area. Huainan has 15.3 billion tons of proved coal reserves and 44.4 billion tons of prospective reserves, accounting for 71% of the total coal reserves of Anhui Province and 32% of that of east China. It is also one of China’s 13 “billion tons of coal and electricity bases.” Figure 1 displays the location map of the study area.

Fig. 1
figure 1

Location map of the study area

Full size image

Panel 16,118 is an underground coal mine panel in Huainan located below the Paifang Village. The strike length, incline length, mining depth, and average depth of the loose layer of the panel are 324, 243.7, approximately 450–500, and 300 m, respectively. The extract height and incline angle of the seam are 2.6 m and 10°, respectively. The mining method used is long-wall fully mechanized mining. The period of extraction was 1 September 2009–20 April 2010. Figure 2 shows the geological section.

Fig. 2
figure 2

Geological section

Full size image

Figure 3 shows the mining plan and the location of focused measuring houses. House A is located on the outer edge of the panel, and houses B and C are located inside the panel. All three houses are near the stop line and are selected as research objects.

Fig. 3
figure 3

Mining plan of panel 16,118 and location of measuring houses A, B, and C

Full size image

Total station and differential level monitoring the houses

Layout of monitoring points and measuring dates

To have a clear understanding of the layout of the measure points around houses A, B, and C, we magnified the layout map, and the layout of the measure points is shown in Fig. 4. Two types of points can be observed: house measure points (in the wall and above ground about 80 cm) and ground measure points (in the ground and distance to the wall about 150 cm).

Fig. 4
figure 4

Layout of house measure points

Full size image

Houses A, B, and C were monitored seven times on 31 December 2009 (panel advanced distance of 73 m), 23 February 2010 (185 m), 18 March 2010 (280 m), 20 April 2010 (324 m), 19 May 2010 (stop), 16 July 2010, and 9 October 2010. The monitoring points lie in each wall of the houses to unify the direction of deformation. The starting points were A11, A12, B11, B12, C11, and C12 (Fig. 4); clockwise was the forward direction; and the calculated direction of horizontal movement was the azimuth along each wall.

Movement and deformation of the houses

Houses A, B, and C are located in three positions in the ground: outside boundary, middle boundary, and inner boundary of panel. Some differences in movement and deformation exist among the houses. The below analysis is based on the data of 9 October 2010, as shown in Table 3. Subsidence was obtained by leveling, horizontal movement was obtained by total station, tilt and curvature were calculated by subsidence, and horizontal strain was calculated by horizontal movement. The maximum subsidence was demonstrated by house C, followed by house B and house A. The largest horizontal movement was exhibited by house B, followed by house C and house A. The maximum tilt was displayed by house B, followed by house C and house A. The maximum curvature was shown by house A, followed by house B and house C. The maximum horizontal strain was depicted by house A, followed by house B and house C.

Table 3 Movement and deformation of house A, B, and C
Full size table

Therefore, the house in the center of the panel showed larger subsidence but smaller deformation than the two other houses. Meanwhile, the house in the inner boundary of the panel was first affected by tensile strain and then by compression strain in the mining processing, and the subsidence of the wall and ground points was similar. The curvature, tilt, and horizontal movement of the wall points were smaller than those of the surface points. The horizontal strain of the houses was smaller than that of the ground, especially the compression strain, where the difference was very large. Based on the observed data, we estimate the subsidence curve of the main section P in dip direction of the panel, which is shown in Fig. 5.

Fig. 5
figure 5

Subsidence profile of section line P in Fig. 3

Full size image

TLS monitoring of the houses

The first scan was conducted on 1 February 2010 when the working face advanced by 87 m and the distance to house C was 165 m. On that date, mining deformation did not affect all the houses. The second scan was performed on 26 April 2010, which was 6 days after mining stoppage.

To obtain the absolute coordinates of the two scans, we set up two control points in the stable area outside the affected mining area. Each scan used control points as the starting coordinates, and the coordinates of the monitoring base point were updated by total station traverse surveying.

To obtain the movement and deformation of the building from the point clouds of the two scans, we selected the corners at the top corners of the building as the measuring points. On the basis of the two scans’ point clouds, the corner feature point at the top of the building was also used as a measuring point. The data of house C were selected as an example, as shown in Table 4, which lists the vector displacement, movement in the x- and y-directions, and subsidence of the measured points. The changes in the line were calculated on the basis of the changes in the monitoring distance between the two measuring points.

Table 4 Movement of measuring points in house C
Full size table

According to the monitoring data of the three houses, although houses B and C are located in the panel, the movement and deformation of house C were the most severe mainly because of the inclination of the coal seams. Moreover, the center of the subsidence basin shifted to the dip direction of the coal seams.

The traditional surveying and TLS for subsidence and horizontal movement are uniform in magnitude and direction. The tilt of the two methods is uniform in direction, but the difference in magnitude is large. In consideration that the feature points lie on top of the house, the tilt is affected by this condition. The difference in horizontal strain is small.

Houses A, B, and C underwent varying degrees of movement and deformation. Their surfaces did not show significant cracks mainly because of the use of structural columns, ring beams, and other strengthening techniques in their construction. The quality of the houses considerably improved during construction. Moreover, no significant cracks were observed on the houses regardless of their stress state. On the basis of the changes in the line of the houses monitored by traditional surveying and TLS, finite-element numerical simulation was performed to evaluate the internal force of the houses in “Analysis of internal forces on houses caused by line changes”.

Analysis of internal forces on houses caused by line changes

The three houses used in this study have the same basic structure, and the authors used a type of general housing structure in the village to analyze the stress state of the houses. The finite-element software ANSYS was utilized to analyze the model, Table 5 lists the material parameters, and Fig. 6 shows the entire model of the front wall. The left half of the model was modeled and analyzed using the plane element with consideration of symmetry to improve the accuracy and efficiency of the calculation. In the model, the structural column, ring beam, foundation ring beam, strip foundation, and wall were considered, and different material properties were defined. The dimensions of each part of the model are as follows (unit: mm): foundation section, 400 × 100; foundation ring beam section, 400 × 400; structural column, 240 × 240; ring beam, 240 × 300; floor height, 3000; axial spacing of structural column, 3300; window, 1500 × 1500; and wall thickness, 240. In the analysis, the influence of tensile deformation on the front wall under deadweight load was considered. Table 6 lists the analysis conditions.

Table 5 Material parameters of the finite-element model
Full size table
Fig. 6
figure 6

Finite-element model of front wall

Full size image
Table 6 Analytical conditions
Full size table

Axial stress of ring beams and structural columns

Figures 7 and 8 shows the axial stresses of the structural columns and ring beams under both conditions, where SX and SY denote the horizontal and vertical principal stresses, respectively. Under the basic condition, the maximum vertical principal stress was concentrated in the middle of the ring beam between the structural columns, and the horizontal principal stress of the second layer was slightly higher than those of the other layers. Under the action of tensile deformation (line changes), the horizontal stress of the foundation ring beam was greatly affected.

Fig. 7
figure 7

Axial stresses of structural columns and ring beams under basic conditions (entire wall)

Full size image
Fig. 8
figure 8

Axial stresses of structural column and ring beam under deformation conditions (half wall)

Full size image

Analysis of main stress results of the wall

The maximum principal and maximum shear stresses of the wall were used to evaluate the risk of wall cracking under the two conditions. Figures 9 and 10 present the principal stresses of the wall under the two conditions and the maximum shear stresses of the wall under the two conditions, respectively. Table 7 shows that the principal and shear stresses increase proportionally with the increase in tensile deformation. The window corner is the principal point of stress for the wall with a window, where the potential risk of cracking is great.

Fig. 9
figure 9

Principal stress of wall

Full size image
Fig. 10
figure 10

Shear stress of wall

Full size image
Table 7 Comparison of results of maximum principal and shear stresses of the wall
Full size table

The foundation ring beam and the inter-story ring beam enhanced the ability of the house to resist the surface movement and deformation, improved the integrity of the house, and reduced the occurrence of cracks, as shown in Fig. 11.

Fig. 11
figure 11

Damaged two-story building with foundation and interlayer ring beam

Full size image

Discussion

The discussion on monitoring results

The houses A, B, and C are respectively located at three positions above the panel, namely, the outer edge, the middle part, and the inner edge of the panel. In this study, through monitoring, house C was first affected by underground mining, and its subsidence was the largest. The horizontal movement of house B is the largest, and the tilt of house C is the largest. Building A is located at the edge of panel (ground tensile deformation zone), houses B and C are located at the inside of panel (ground compression deformation zone), and the compression value of house B is larger than that of house C.

Therefore, the general law is that the subsidence of the building in the middle of the panel is larger, but the other deformation values are relatively small. In the mining process, the building is first affected by tensile deformation and then by compression deformation. The subsidence of the building is basically consistent with the subsidence of the ground. Curvature, tilt, and horizontal movement of buildings are smaller than the ground, but the difference is small. The horizontal deformation of buildings is smaller than that of the ground, and the difference is great, especially for the horizontal compression deformation.

The discussion on numerical simulation results

As shown in the numerical simulation section in this paper, when there is no sliding layer of the building foundation, the deformation of the wall is transitive under the action of tensile deformation, which has a great influence on the foundation ring beam and the bottom layer. With the increase of tensile deformation, the principal stress and shear stress of the wall increase proportionately. When there is a window hole, the corner of the window is the key point of stress, and the potential cracking risk is large.

When there is a sliding layer in the building foundation, the sliding layer will reduce the upward transfer of the surface deformation, which greatly reduces the axial stress of the structural column and the foundation ring beam, especially the absorption effect on the tensile deformation. Because the shear slip of the sliding layer consumes the bottom tension of the foundation, the transfer path of deformation is effectively cut off, so that the deformation of the foundation ring beam and the wall is small, the influence of tensile deformation on the stress of the wall is reduced, and the cracking of the wall under tensile condition is effectively slowed down.

The highlights of this study are that the combination of ground monitoring and three-dimensional laser scanning can monitor the building in an all-round way, and the monitoring results can be used to simulate and analyze the internal movement and deformation of the building, so as to determine the internal movement and deformation of the building in the mining area.

The Limitations and future research are: (1) external monitoring by TLS or traditional surveying methods is expensive. Many houses are located in the coal-mining subsidence area, but only three houses are monitored because of monitoring means and financial constraints. The InSAR technique can be used in the future to monitor the deformation of the disturbed houses in real-time and all-weather conditions. (2) The numerical simulation for obtaining the internal mechanical state of the disturbed houses in this study is a general quantitative case, which is based on the monitor results of the three houses. In the future, an automatic system should be developed to combine the deformation results from InSAR and, thus, diagnose the disturbed houses’ internal state.

The connection of monitoring and numerical simulation

The construction of the coal-mining subsidence area is easy to observe the external deformation affected by mining, and the stress state of the building and structure is difficult to carry out no damage monitoring. Therefore, the deformation monitoring of the first part of this paper provides the working condition data of building and structure for the numerical modeling of the second part. Particularly horizontal deformation, curvature are sensitive to the building and structure.

The assumption of numerical simulation should be used

In real life, if the quality of the building in the coal-mining subsidence area is good enough, although the building is affected by non-uniform deformation, there is no damage, and the surrounding houses have been cracked or destroyed; at this time, it is necessary to use numerical simulation to evaluate the internal stress state of the exterior intact building to protect it in advance.

Conclusions

  1. 1.

    The traditional surveying (total station and level) and TLS were utilized to monitor the movement and deformation of the disturbed houses affected by underground coal mining. The monitoring results of these two methods are consistent. The external monitoring of the disturbed houses supplies the reference data for the next evaluation of internal force state of the houses by numerical simulation. The influence of horizontal deformation on buildings is more serious, in which the influence of tensile deformation is much greater than that of compression deformation, while synchronous subsidence and movement have no effect on buildings, and tilt has great influence on tall buildings, but relatively little effect on buildings with lower height.

  2. 2.

    Numerical simulation shows that under the action of tensile deformation, the structure of a damaged building in a subsidence zone exerts a considerable influence on the foundation ring beam. An increase in tensile deformation proportionally increases the principal and shear stresses of the wall. The window corners are the principal points of stress when a window exists in the wall, and they are the most at risk for cracking. The brick–concrete structure with foundation sliding layer, reinforced concrete ring beam, and structural column has the best ability to resist surface deformation, and the sliding layer plays an important role in reducing the horizontal deformation of the ground.

  3. 3.

    The combination of external monitoring and internal numerical simulation provides a new way to evaluate the damaged degree of the affected houses by underground mining.