Application of electrical resistivity imaging to detection of hidden geological structures in a single roadway

2.1 Basic principles

The parallel coal seam detection method arranges observation systems in coal seams to collect data, Figure 1(a). Because of the shielding of current by the roadway cavity, the majority of supply current flows into the working face, and the potential difference measured primarily contains information on geoelectric anomalies in the working face. In practice, two electric parameters from electrical resistivity imaging methods are generally calculated using equation (1) according to the potential difference ΔUMN between the electrodes and the relative positions of the electrodes [26,27,28].

where ΔUMN denotes the potential difference between the electrodes, mV; I denotes the supply current intensity, mA; K denotes the array coefficient.

Figure 1

(a) Schematic diagram of the observation system layout and (b) a three-layer geoelectric model.

In coal measures, coal seams and their roof and floor strata are composed of layered mediums with different electric properties. For the simple three-layer geoelectric model shown in Figure 1(b), the potential expression of a point source in full space can be solved using the image method:

where (r,z) denotes the coordinates of the measuring point in cylindrical coordinates; i denotes the number of medium in which a point source is located; j denotes the number of medium in which an observation point is located; aij,bij,cij denote undetermined coefficients, which can be solved using boundary conditions; λ is an integral variable; J0(λr) is a zero-order Bessel function of the first kind.

Point sources are located in the middle of seam floor and coal seam, respectively, and equation (2) is used to calculate the potential, as shown in Figure 2. It can be known that: (1) the potential isoline distorts the interface of coal strata, but still maintains its continuity. (2) When a field source is located on the seam floor, electric field energy is primarily concentrated in the floor surrounding rock; when a field source is located in the middle of coal seams, the potential isoline presents an approximately elliptical distribution, and more current flows to coal seams. Adopting consequent detection methods obtains more information on geoelectric anomalies in coal seams.

Figure 2

Distribution of the point electric field. (a) Source located in the coal seam floor and (b) source located in the middle of the coal seam.

2.2 Observation system design

General working faces are less than 200 m in width (L < 200 m), and, within this prospecting scope, electrical resistivity imaging methods have a high resolution; hence, consequent detection methods can be used to realize the detailed detection of concealed geological structures in “single-roadway” working faces. In the observation system design for the consequent detection method, a close attention should be paid to the following aspects:

1. Array mode: it is required to guarantee sufficient sensitivity and minimize the “blind area” in prospecting.

2. Measuring line layout: measuring lines should be arranged with reference to exposed geological structures, and measuring line length should be greater than the length of the target prospecting section.

3. Parameter selection: polar distance sequence should be determined according to prospecting depth and roadway supporting conditions, guaranteeing both the precision of shallow detection and the accuracy of deep measurement.

4. Interference suppression: in practical construction, it is necessary to suppress the interference caused by bodies with heterogeneous electric properties (such as the anchor net and anchor rod) and electrode earthing resistance, ensuring signal quality.

3.1 Geoelectric model establishment

In the primary state, the sedimentary sequence of coal strata is clear, and geoelectric characteristics follow regular variation laws in the longitudinal direction (generally manifested as a resistivity ranking of mudstone < sandstone < coal seam). In the transverse direction, resistivity distribution is relatively homogeneous. In cases where the geological or groundwater movement causes stratigraphic dislocation, erosion, or caving and forms fault fissure zones, karst caves, collapse columns, or other concealed geological structures, the intrinsic electric variation laws of primary coal measure strata are altered, whether water still present or not. Considering that concealed geological structures commonly found in coal seams are typically collapse columns, they can be abstracted as cuboids in full-space infinitely large mediums, as shown in Figure 3.

Figure 3

Schematic diagram of the collapse column geoelectric model and line layout.

Geoelectric models are 600 m (length) × 200 m (width) × 200 m (height) in size. Collapse columns penetrate coal seams, and are developed in both the roof and floor and are 40 m × 40 m × 100 m in size. Measuring lines are arranged in the middle of coal seams. In addition, this paper sets coal seam thickness (roadway height) as h, surrounding rock resistivity as ρw, coal seam resistivity as ρm, and roadway resistivity as ρh.

3.2 Influence correction of the roadway cavity

Given that the parallel coal seam detection method is used in the mine roadway construction, the presence of roadway cavities is bound to cause distortion in full-space steady current fields. In this sense, studying methods for correcting the influence of roadway cavities is vital for data processing and interpretation using consequent detection methods.

For the sake of generality, this paper makes the following assumptions: roof and floor surrounding rock resistivity ρw=200Ωm, coal seam resistivity ρm=500Ωm, roadway cavity resistivity ρh=1.0×1011Ωm, roadway height h=5m, and roadway width a=1,2,…,9,10m. A 3D finite element method is used for the forward calculation of geoelectric models of different roadway widths, as shown in Figure 4(a). It can be known that: the roadway cavities have a significant influence on the small polar distance; with increasing polar distance, the influence of roadway cavities on measurement results presents an exponential decay, and apparent resistivity gradually approaches surrounding rock resistivity; with increasing roadway width (section), the influence scope of roadway cavities gradually expands.

Figure 4

Influence correction of the roadway cavity. (a) Gray-scale map of apparent resistivity for different roadway widths and (b) apparent resistivity curves.

To further study methods of correcting the influence of roadway cavities, this paper assumes the relationship between apparent resistivity and full-space apparent resistivity under the influence of roadway cavity as:

where ρsa denotes the full-space apparent resistivity; ρsm denotes the apparent resistivity under the influence of roadway cavities; KD denotes the correction coefficient, which is a function of the roadway cross-section area and electrode spacing; S denotes the roadway cross-section area; L denotes the electrode spacing.

In practical applications, electric parameters of surrounding rock and coal seams can be measured in the field and used to establish geoelectric models with/without roadways for the forward calculation of background fields. Next, a ratio operation is performed to obtain the correction coefficient. Finally, full-space apparent resistivity is acquired from the measured apparent resistivity data. Figure 4(b) shows the apparent resistivity curve of the geoelectric model for low-resistivity collapse columns when the roadway width is a=5m. Clearly, adopting the correction method from equation (3) can effectively suppress the influence of roadway cavities on the small polar distance, and the curve after the roadway cavity correction basically overlaps with the curve without roadway cavities; however, with increasing polar distance, the curve after correction gradually overlaps with the curve with roadway cavities, meaning that the correction method cannot eliminate the influence of roadway cavities on the anomaly response amplitude.

Parallel coal seam detection method arrange observation systems in the middle of coal seams on the roadway side, and roadway cavities (high-resistivity layer) can be seen as a current shielding layer, causing more current to flow into coal seams. Thus, roadway cavities may influence the anomaly response amplitude. Figure 5 shows the apparent resistivity variation rates for a Schlumberger array when collapse columns are located in coal seams and on the outside of the roadway (calculated from equation (4)), respectively. By comparing Figure 5(a) and (b), the shielding of current by roadway cavities increases the anomaly response amplitude in coal seams and weakens the influence of anomalous bodies on the outside of roadway on observation results. That is, roadway cavities enhance the ability of the parallel coal seam detection method to detect concealed geological structures in working faces.

where η denotes the apparent resistivity variation rate; ρsabnorm denotes the apparent resistivity with anomalous bodies; ρsnorm denotes the apparent resistivity of background field; x denotes the observation point position.

Figure 5

Apparent resistivity section curves of the Schlumberger array (OA = 100 m). (a) Abnormal body located in the coal seam and (b) abnormal body located outside the roadway.

5.1 Construction design

During the cut-hole excavation from conveyor roadway to coal seam in the 92,606 working face of the mine in Jiangsu Province, a collapse column about 80 m away from the conveyor roadway was found. The collapse column was characterized by missing coal seam, incomplete lithology, and loose filler. Neither water burst nor water drenching was present. The preliminary judgment was that the collapse column was a non-water-flowing and non-water-bearing collapse column.

The 92,606 working face was about −750 m in elevation and 160 m in width; airway caved during the stoping of the adjacent working face; cut-hole had a high slope and lacked adequate support. To guarantee detection precision and construction safety, a consequent pole–dipole array detection system was designed, as shown in Figure 10. Fifty-one electrodes were arranged at cut-hole positions from left to right with an electrode spacing of 5 m (measuring line length = 250 m) and were connected with each other using a communication line; “infinite” electrodes were arranged about 1,000 m way from cut-holes along the roadway. Because of the high resistivity of coal seams and coal gangues, the electrode earthing resistance was high. To ensure the effectiveness of collected signals, holes 20–30 cm in depth were drilled at electrode locations, and wet clay with an excellent electrical conductivity was stuffed inside.

Figure 10

Schematic diagram of the resistivity observation system of the 92,606 working face.

5.2 Analysis of measured data

Data were collected using the parallel electrical survey-based system. Each electrode in the system was always kept in a voltage sampling state, except for the power supply; measured data, collected in large volumes, were transmitted to the host via the communication line in real time.

After the conversion of field data, pole–dipole array data were extracted; on the basis of field measured data (roadway width = 6 m; roadway height = 5 m; coal seam resistivity = 700 Ω m, and surrounding rock resistivity = 250 Ω m), a geoelectric model was established, and the influence correction coefficient of the roadway cavity was obtained through a forward calculation. Measured data, after being corrected (Figure 11(a)), were put through smoothing and interpolation processing, and apparent resistivity isolines were drawn, Figure 11(b). Experimental measurements of the pole–dipole array are given in Figure 11(b). These results revealed that, at a longitudinal depth of 20–30 m, the apparent resistivity was generally low, which was conjectured to be a consequence of the roadway floor ponding and not interpreted as an anomaly. At a transverse position of 5–60 m and a longitudinal depth of 80–150 m (area demarcated by blue dotted lines), the apparent resistivity was high, which was conjectured to be a high anomaly response induced by non-water-flowing and non-water-bearing collapse columns. At a transverse position of 80–100 m and a longitudinal depth of 30–70 m (area demarcated by red dotted lines), the apparent resistivity was low and presented a zonal distribution, which was conjectured to be a fault. The detection results fit well with the drilling verification results obtained by the mine owner and deviated from the actual collapse column position by about 3 m, which could meet the production demands of the mine owner.

Figure 11

Observation results of the pole–dipole array. (a) Apparent resistivity curves (x = 47.5 m) and (b) apparent resistivity isolines.