Roadway roof instability risk identification method based on while-drilling detection technology

Abstract

The invention discloses a roadway roof instability risk identification method based on a while-drilling detection technology. The method comprises the following steps that parameters such as thrust, torque and rotating speed in the drilling process are collected in real time; establishing a rock stratum strength prediction model by a multiple linear regression method, and determining lithology and rock sequence distribution of a roof rock stratum; analyzing and predicting the intensity sequence by using an accumulative sum algorithm, identifying a rock stratum mutation point, and judging a weak intercalated layer and a fracture development region; calculating four instability risk evaluation indexes such as a roof loosening circle range and a self-stabilization balance arch height by combining rock stratum strength and structural characteristics; and fusing multi-index data through a D-S evidence theory model, dividing five risk levels and generating a risk partition cloud picture. According to the method, the problems of poor real-time performance, difficulty in dynamic risk quantification and insufficient visualization of a traditional method are solved, geological information can be dynamically monitored in real time, and a hidden structure can be accurately identified; multi-source data fusion is combined with a visual cloud picture, the risk probability is scientifically quantified, and a visual basis is provided for underground safety decision making.

Description

Technical Field

[0001] The present invention relates to the field of intelligent coal mining, and in particular to a method for identifying the risk of roadway roof instability based on detection-while-drilling technology. Background Art

[0002] As coal mining depths continue to increase, tunnel roof instability has become a core threat to mine safety. Currently, the industry generally uses manual core sampling combined with laboratory mechanical testing to assess rock formation strength. This method has significant limitations: First, traditional drilling and coring takes hours or even days, making it impossible to obtain dynamic geological information about the excavation face in real time, causing support solutions to lag behind actual working conditions. Second, laboratory testing can only reflect the static mechanical properties of local rock samples and is unable to reveal the spatial distribution of hidden structures such as crack development and weak interlayers within the rock formation.

[0003] Existing methods for identifying roof instability primarily rely on empirical formulas or static weight evaluation systems, which have exposed two major technical bottlenecks in engineering practice: First, traditional evaluation models struggle to quantify the inherent correlations between dynamic risk factors, such as sudden changes in rock strength and the evolution of structural fragmentation, and lack an effective fusion decision-making mechanism when faced with conflicting multi-indicator data. Second, existing monitoring systems often output results in the form of numerical reports, making it difficult for underground workers to obtain timely and intuitive information on the spatial distribution of risks, seriously affecting emergency response efficiency. These issues have severely restricted the advancement of intelligent tunneling and safety management in coal mine tunnels. There is an urgent need to develop a roadway roof instability risk identification method that integrates real-time perception, intelligent inversion, and visual decision-making. Summary of the Invention

[0004] The purpose of the present invention is to provide a method for identifying the risk of tunnel roof instability based on while-drilling detection technology, which integrates real-time perception, intelligent inversion and visual decision-making, and can realize the automation of the entire process of "monitoring-analysis-early warning".

[0005] In order to achieve the above-mentioned object of the invention, the present invention adopts a technical solution as follows: the present invention provides a method for identifying the risk of instability of the roadway roof based on the detection while drilling technology, comprising the following steps:

[0006] S1: The drilling parameters during the drilling process are collected in real time through the drilling parameter measurement instrument installed on the drilling rig, including thrust, torque, speed, displacement, drilling speed and vibration;

[0007] S2: Based on the drilling parameters collected in step S1, the comprehensive indicators SEM and RDA are calculated, and a rock formation strength prediction model is established through a multivariate linear regression method to determine the lithology and rock sequence distribution of the roof rock formation;

[0008] S3: Input the predicted rock strength data into the cumulative sum algorithm CUSUM to obtain the mutation points in the predicted strength data sequence, and determine the weak interlayer and crack development areas based on the mutation points;

[0009] S4: Based on the rock formation strength obtained in step S2 and the rock formation structure identification results obtained in step S3, four instability risk assessment indicators are calculated, namely, the range of the roof loosening zone, the height of the roof self-stabilizing balance arch, the amount of roof separation at the weak interlayer position, and the direct roof stress intensity index;

[0010] S5: Input the four instability risk assessment indicators into the DS evidence theory model for multi-source information fusion, calculate the tunnel roof instability risk probability at each drilling location, and generate a tunnel roof instability risk zoning cloud map.

[0011] Preferably, in step S1, the drilling parameters during the drilling process are collected in real time by a drilling parameter measuring instrument installed on the drilling rig, wherein a thrust sensor, a torque sensor, a speed sensor, a displacement sensor and a vibration sensor are installed inside the drilling parameter measuring instrument;

[0012] Preferably, step S2 specifically includes the following steps:

[0013] S21: The calculation formulas for the two comprehensive indicators SEM and RDA using the drilling parameters are as follows:

[0014]

[0015] Where F is thrust, T is torque, v is drilling speed, w, c, a, ka, C, λ, α, β, and γ are preset parameters, among which w is used to enhance the weak signal in the drilling specific energy sequence; c specifies the switching point of the logic function; a controls the transition interval of the SEM curve when the lithology changes; k a and C are constants respectively; λ, α and β are important parameters controlling the functional relationship between rock strength and drilling parameters;

[0016] S22: The multivariate linear regression equation for predicting rock formation strength using SEM and RDA, i.e., the rock formation strength prediction model, is expressed as:

[0017] R c =k1×SEM+k2×RDA

[0018] Among them, K1 is the regression coefficient of SEM in the regression equation; K2 is the regression coefficient of RDA in the regression equation.

[0019] Preferably, the specific method of step S3 is:

[0020] S31: Assume that the amount of predicted intensity sequence data of a certain borehole is n, and the drilling parameter sequence is divided into m intervals, then the number of drilling parameters in each interval is q = n / m;

[0021] S32: Calculate the coefficient of variation Vc of the predicted strength of each interval:

[0022]

[0023] Where V ci is the coefficient of variation of the i-th drilling parameter interval, σ i and are the standard deviation and mean of the i-th drilling parameter interval, is the jth data point in the i-th interval;

[0024] S33: recombining the calculated coefficients of variation of each interval into a new coefficient of variation sequence in order, where the number of data in the coefficient of variation sequence is q;

[0025] S34: Set the minimum error e of the cumulative sum algorithm CUSUM r Sum the cumulative sum and initial value s0, calculate the coefficient of variation data sequence i data point V ci and the mean coefficient of variation The deviation value d i ;

[0026]

[0027] S35: Accumulate the deviation value d in sequence according to the sequence of the coefficient of variation i , get the new cumulative sum of the coefficient of variation sequence s i :

[0028]

[0029] S36: Calculate the coefficient of variation V ci The corresponding cumulative sum threshold t is continuously iteratively optimized according to the existing data characteristics during the calculation process to suppress the interference of data fluctuations on the change point monitoring results. The threshold iteration model is as follows:

[0030]

[0031] Where μ represents the characteristic parameter between the monitoring threshold and the coefficient of variation;

[0032] S37: When s i When t > t, the interval fluctuation degree is considered to exceed the threshold, and the interval is marked as a change point, indicating the location where the rock structure suddenly changes, and the interval number i corresponding to the value is output;

[0033] S38: Assume that there are x intervals between the pth change point and the p+1th change point, calculate the mean of the rock strength in each interval, and compare the maximum strength with the set strength threshold. When the predicted strength is lower than the minimum threshold, it means that there is crack development between the two change points. When the strength is between the maximum and minimum thresholds, it means that there is a weak interlayer between the two change points. When the strength is higher than the maximum strength threshold, it means that there is a roof rock layer between the two change points.

[0034] Preferably, in step S4,

[0035] The calculation equation for the range of the top plate loose circle is as follows:

[0036]

[0037] Where: a is the roadway radius, m is the lithology-related constant, σ c is the uniaxial compressive strength of the immediate roof, s is the constant related to the degree of roof crushing, P0 is the original vertical stress of the roadway, and R0 is the range of the roof loosening zone;

[0038] The calculation equation for the height of the self-stabilizing balance arch of the roof is as follows:

[0039]

[0040] Where: f is the rock strength coefficient of the immediate roof, λ is the lateral pressure coefficient, 2a is the tunnel width, f c is the mining impact coefficient;

[0041] The calculation equation for the roof separation at the weak interlayer position is as follows:

[0042]

[0043] Where: η i =√P i / E i I i is the horizontal force at the end of the i-th layer rock beam; E i is the elastic modulus of the i-th rock beam; I i is the cross-sectional inertia moment of the i-th layer of rock beam; q i is the vertical load on the i-th layer of rock beam, γ i is the bulk density of the i-th layer of rock beam; h i is the thickness of the i-th rock beam, w i, is the subsidence of the i-th layer of rock beam, l i is the span of the i-th rock beam, p i Horizontal force at the fixed end of the i-th layer rock beam, η i =p i / E i I i ;

[0044] The calculation equation of the direct top stress intensity index is as follows:

[0045]

[0046] Where: k d is the balance coefficient, σ c is the uniaxial compressive strength of the direct top, h d is the thickness of the direct roof, σ H is the vertical principal stress of the tunnel surrounding rock, R c It is the strength of the immediate top rock layer.

[0047] Preferably, the specific method of step S5 is:

[0048] S51: The roadway roof instability risk level is divided into five levels: T1 (I) safe, T2 (II) basic safe, T3 (III) critical safe, T4 (IV) basic dangerous, and T5 (V) dangerous. The value ranges of each evaluation index under different instability risk levels are shown in the table:

[0049] Evaluation index range under different instability risk levels

[0050]

[0051] S52: The power set within the comprehensive assessment framework for instability risk contains 32 elements:

[0052]

[0053] S53: For the recognition framework Θ, the quality function of any subset A is its power set 2 Θ The mapping to [0,1] is represented by m:2 Θ →[0,1], m is the quality function of Θ on the recognition frame, and its calculation formula is as follows:

[0054]

[0055] Where: x is the characteristic value of the evaluation index, Identify the center of the interval for each evaluation indicator, and are the upper and lower limits of each identification interval respectively; when When , the membership degree reaches the maximum value of 1, indicating a state of complete trust;

[0056] S54: Definition The trust function is Bel:2 Θ →[0,1], Bel(A) represents the total trust level of the evidence on subset A, which constitutes the lower bound of the probability distribution of subset A; the trust function satisfies the following conditions:

[0057]

[0058] S55: Definition The likelihood function is Pl:2 Θ →[0,1], Pl(A) constitutes the upper limit of the probability distribution of subset A; the likelihood function satisfies the following conditions:

[0059]

[0060] S56: Assume m1, m2, ..., m N Corresponding to A1, A2, ..., A N The quality function of , then the synthesis formula of N pieces of evidence information is as follows:

[0061]

[0062] Where: ⊕ represents the orthogonal sum, K is the evidence conflict coefficient, the larger the K is, the greater the conflict between the two evidences is, and the condition for the two evidence indicators to be fused through the Dempster combination rule is 1 <K;

[0063] S57: Substitute the instability risk evaluation index value into the DS evidence theory evaluation model to obtain the roof instability risk probability of each drilling location, and draw the instability risk zoning cloud map of different areas of the tunnel roof.

[0064] Compared with the prior art, the present invention has the following beneficial effects:

[0065] 1. Real-time geological data collection using drilling-while-drilling technology (e.g., parameters such as thrust, torque, and rotational speed) eliminates the time-consuming bottleneck of traditional coring and enables immediate feedback on rock formation characteristics during excavation. A rock formation strength prediction model is established using comprehensive indicators (SEM, RDA) and multivariate linear regression methods to dynamically predict rock formation strength and rock sequence distribution, reducing reliance on static laboratory testing.

[0066] 2. The cumulative sum algorithm (CUSUM) is used to analyze and predict the intensity sequence, effectively capturing the mutation points of the rock formation, accurately identifying hidden structures such as weak interlayers and crack development, and improving the accuracy of geological risk identification. Through iterative optimization of the coefficient of variation and threshold, data fluctuation interference is suppressed, and the dynamic tracking capability of the evolution of rock formation fragmentation is enhanced.

[0067] 3. Based on the DS evidence theory model, four instability risk assessment indicators—roof loosening zone range, roof self-stabilizing arch height, roof delamination at weak interlayer locations, and direct roof stress intensity index—are integrated to resolve multi-indicator conflicts. Risk probability is quantified through orthogonal and rule-based approaches, enhancing the credibility of identification results. Five risk levels (from safe to dangerous) are assigned, and the final risk level is determined based on the principle of maximum support, providing a scientific and hierarchical basis for decision-making.

[0068] 4. Generate an instability risk zoning cloud map, intuitively display the risk level and probability of different areas through color gradients, mark high-risk drilling locations, and greatly improve the emergency response efficiency of underground personnel. BRIEF DESCRIPTION OF THE DRAWINGS

[0069] The accompanying drawings are used to provide further understanding of the present invention and constitute a part of the specification. They are used to explain the present invention together with the embodiments of the present invention and do not constitute a limitation of the present invention.

[0070] Figure 1 A flow chart of a method for identifying the risk of roadway roof instability based on detection-while-drilling technology provided by the present invention;

[0071] Figure 2 A schematic structural diagram of the data acquisition system provided by the present invention;

[0072] Figure 3 A schematic diagram of the data acquisition process of the data acquisition system provided by the present invention;

[0073] Figure 4 The collected while-drilling parameter map provided by the present invention;

[0074] Figure 5 The relationship curve between the rock formation strength and the two indicators of SEM and RDA provided by the present invention;

[0075] Figure 6 This is the rock structure diagram identified using the CUSUM method;

[0076] Figure 7 It is the probability distribution diagram of evaluation indicators;

[0077] Figure 8 The probability map of different instability risk levels of ZK2 borehole obtained using the DS model;

[0078] Figure 9 The results of evaluating the instability risk level of the roadway within the range of 300 to 570 m using DS drilling are presented. DETAILED DESCRIPTION

[0079] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention will be further described in detail below in conjunction with the accompanying drawings and embodiments. Of course, the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0080] See also Figure 1 This embodiment provides a method for identifying the risk of roadway roof instability based on detection while drilling technology, comprising the following steps:

[0081] S1: Use the data acquisition system to collect drilling parameters during the drilling process, including thrust, torque, speed, displacement, drilling speed and vibration;

[0082] The data acquisition system includes a downhole parameter measuring instrument installed on the drilling rig, which collects downhole parameters in real time during the drilling process. The downhole parameter measuring instrument 2 is internally installed with a thrust sensor, a torque sensor, a speed sensor, a displacement sensor and a vibration sensor. The lower end of the downhole parameter measuring instrument 2 is connected to the single-body anchor drilling rig using an external hexagonal shaft, and the upper end is connected to the B19 hexagonal drill rod 7.

[0083] See also Figure 2-Figure 3 The device used in this step also includes a signal wireless transmission module 3, a signal wireless receiving module 4, a multi-channel data acquisition module 5, and a data storage and display terminal 6. The drilling rig uses a single anchor drilling rig 1. The thrust sensor, torque sensor, speed sensor, displacement sensor, and vibration sensor transmit the collected data to the signal wireless receiving module 4 through the signal wireless transmission module 3, and then transmit it to the multi-channel data acquisition module 5 and the data storage and display terminal 6;

[0084] S2: Based on the drilling parameters collected in step S1, the comprehensive indicators SEM and RDA are calculated, and a rock formation strength prediction model is established through a multivariate linear regression method to determine the lithology and rock sequence distribution of the roof rock formation;

[0085] S3: Input the rock strength data predicted by the rock formation strength prediction model into the cumulative sum algorithm CUSUM to obtain the mutation points in the predicted strength data sequence, and determine the weak interlayer and crack development areas based on the mutation points;

[0086] S4: Based on the rock formation strength obtained in step S2 and the rock formation structure identification results obtained in step S3, four instability risk assessment indicators are calculated, namely, the range of the roof loosening zone, the height of the roof self-stabilizing balance arch, the amount of roof separation at the weak interlayer position, and the direct roof stress intensity index;

[0087] S5: Input the four instability risk assessment indicators into the DS evidence theory model for multi-source information fusion, calculate the tunnel roof instability risk probability at each drilling location, and generate a tunnel roof instability risk zoning cloud map.

[0088] For specific examples, see Figure 4-Figure 6It can be seen that the overall rotation torque range of the 1st to 2nd drill rods (0-1.5m) is between 4.5 and 10.0 N·m, with an average torque of 7.2 N·m. The rotation torque range of the 2nd to 5th drill rods (1.5-5.2m) is concentrated between 10.0 and 16.0 N·m, with an average torque of approximately 14.6 N·m. The rotation torque range of the 5th to 6th drill rods (5.2-7.0m) is 14.0-22.0 N·m, with an average torque of approximately 18.2 N·m. This shows that the rock formations within 7.0m of the roof at ZK2 are roughly divided into three types.

[0089] The CUSUM algorithm is used to identify potential rock formation mutation points in the data sequence to achieve intelligent identification of the rock sequence and structural distribution of the roof rock formation at the ZK2 drilling location.

[0090] Based on the identification of geological information such as rock strength and rock structure of ZK2 borehole using the drilling parameters of the borehole, the identified rock geological information parameters are used with the instability risk evaluation index calculation equation to calculate the four evaluation index values corresponding to the ZK2 borehole position, namely the range of the composite roof loosening zone, the height of the roof self-stabilizing balance arch, the roof delamination amount at the weak interlayer position, the roof delamination amount at the weak interlayer position, and the direct roof stress intensity index. The values of each index are shown in Table 1 below.

[0091] Table 1: Roof instability risk assessment index values

[0092]

[0093] According to the tunnel instability risk evaluation index values and evaluation index value ranges shown in Table 1, they are substituted into the quality function calculation equation in S53 to obtain the probability distribution of the evaluation results for different evaluation indicators, as shown in the following example: Figure 7 As shown in Figure 2, it can be seen that there are obvious differences in the probability distribution of different evaluation indicators for each instability risk level, and the mean probability distribution of different instability risk levels is calculated respectively.

[0094] According to the established five-level risk classification criteria for roadway roof instability, namely T1 (I) safe, T2 (II) basic safe, T3 (III) critical safe, T4 (IV) basic dangerous, and T5 (V) dangerous, the average probability distribution for level I was 5.9%, the average probability distribution for level II was 17.9%, the average probability distribution for level III was 42.6%, the average probability distribution for level IV was 24.3%, and the average probability distribution for level V was 9.3%. Therefore, based on the mean probability distribution, the roof instability risk level at the ZK2 borehole location is level III, indicating that the roof is in a critical safe state.

[0095] Through the above analysis, the probability distribution of different evaluation indicators for each instability risk level is obtained. On this basis, this paper uses the DS evidence theory to fuse the probability distribution given by each evaluation indicator to achieve a comprehensive judgment of the roof instability risk at the ZK2 drilling position. Figure 8 It can be seen that among the roof instability probabilities at the ZK2 borehole location obtained based on the DS evidence theory, the probability of tunnel roof instability risks at levels II and III is significantly higher than that of other levels. Among them, the probability of level III instability risk is the highest, at 65.1%, and the probability of level II instability risk is 29.1%. Based on the maximum support rate principle, this paper obtains the roof instability risk level of the ZK2 borehole location as level III, which means that the roof is in a critical safety state. This is consistent with the instability risk level obtained using the average probability distribution. However, the instability risk probability given by the DS evidence theory is more prominent and has higher credibility than the average probability distribution.

[0096] Using the above-mentioned rock formation strength and rock formation structure while drilling identification method, the rock formation characteristics corresponding to ZK1 to ZK10 are identified respectively, and the rock formation information while drilling identification results of each borehole are obtained. The rock formation geological identification results are then brought into the instability risk evaluation index calculation equation to obtain the different borehole instability risk evaluation index values shown in Table 2. Table 2: Different borehole instability risk evaluation index values are as follows:

[0097]

[0098] Then, the probability distribution of each drilling evaluation index is calculated using the quality function, and finally the DS evidence theory is used to synthesize various indicators to obtain the tunnel instability risk level in different drilling areas. Figure 9 As shown in Table 2 above, the instability risk level in the range of 300-360m is Level III, the instability risk probability in the range of 390-450m is Level II, the instability risk in the range of 460-500m is between Level II and Level III, and the instability risk in the range of 500-570m is mainly Level III.

[0099] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for identifying the risk of roadway roof instability based on detection while drilling technology, characterized in that: The following steps are involved: S1: The drilling parameters during the drilling process are collected in real time through the drilling parameter measurement instrument installed on the drilling rig, including thrust, torque, speed, displacement, drilling speed and vibration; S2: Based on the drilling parameters collected in step S1, the comprehensive indicators SEM and RDA are calculated, and a rock formation strength prediction model is established through a multivariate linear regression method to determine the lithology and rock sequence distribution of the roof rock formation; S3: Input the predicted rock strength data into the cumulative sum algorithm CUSUM to obtain the mutation points in the predicted strength data sequence, and determine the weak interlayer and crack development areas based on the mutation points; S4: Based on the rock formation strength obtained in step S2 and the rock formation structure identification results obtained in step S3, four instability risk assessment indicators are calculated, namely, the range of the roof loosening zone, the height of the roof self-stabilizing balance arch, the amount of roof separation at the weak interlayer position, and the direct roof stress intensity index; S5: Input the four instability risk assessment indicators into the DS evidence theory model for multi-source information fusion, calculate the tunnel roof instability risk probability at each drilling location, and generate a tunnel roof instability risk zoning cloud map.

2. The method for identifying the risk of roadway roof instability based on detection while drilling technology according to claim 1, characterized in that: In step S1, the drilling parameters during the drilling process are collected in real time by a drilling parameter measuring instrument installed on the drilling rig, wherein the drilling parameter measuring instrument is internally installed with a thrust sensor, a torque sensor, a speed sensor, a displacement sensor and a vibration sensor.

3. The method for identifying the risk of roadway roof instability based on detection while drilling technology according to claim 1, characterized in that: Step S2 specifically includes the following steps: S21: The calculation formulas for the two comprehensive indicators SEM and RDA using the drilling parameters are as follows: Where F is thrust, T is torque, v is drilling speed, w, c, a, ka, C, λ, α, β, and γ are preset parameters. Among them, w is used to enhance the weak signal in the drilling specific energy sequence; c specifies the switching point of the logistic function; a controls the transition interval of the SEM curve when the lithology changes; ka and C are constants; λ, α, and β are important parameters that control the functional relationship between rock strength and drilling parameters. S22: The multivariate linear regression equation for predicting rock formation strength using SEM and RDA, i.e., the rock formation strength prediction model, is expressed as: R c =k1×SEM+k2×RDA Among them, K1 is the regression coefficient of SEM in the regression equation; K2 is the regression coefficient of RDA in the regression equation.

4. The method for identifying the risk of roadway roof instability based on drilling detection technology according to claim 1, characterized in that: The specific method of step S3 is: S31: Assume that the amount of predicted intensity sequence data of a certain borehole is n, and the drilling parameter sequence is divided into m intervals, then the number of drilling parameters in each interval is q = n / m; S32: Calculate the coefficient of variation Vc of the predicted strength of each interval: Where V ci is the coefficient of variation of the i-th drilling parameter interval, σ i and are the standard deviation and mean of the i-th drilling parameter interval, is the jth data point in the i-th interval; S33: recombining the calculated coefficients of variation of each interval into a new coefficient of variation sequence in order, where the number of data in the coefficient of variation sequence is q; S34: Set the minimum error e of the cumulative sum algorithm CUSUM r Sum the cumulative sum and initial value s0, calculate the coefficient of variation data sequence i data point V ci and the mean coefficient of variation The deviation value d i ; S35: Accumulate the deviation value d in sequence according to the sequence of the coefficient of variation i , get the new cumulative sum of the coefficient of variation sequence s i : S36: Calculate the coefficient of variation V ci The corresponding cumulative sum threshold t is continuously iteratively optimized according to the existing data characteristics during the calculation process to suppress the interference of data fluctuations on the change point monitoring results. The threshold iteration model is as follows: Where μ represents the characteristic parameter between the monitoring threshold and the coefficient of variation; S37: When s i When t > t, the interval fluctuation degree is considered to exceed the threshold, and the interval is marked as a change point, indicating the location where the rock structure suddenly changes, and the interval number i corresponding to the value is output; S38: Assume that there are x intervals between the pth change point and the p+1th change point, calculate the mean of the rock strength in each interval, and compare the maximum strength with the set strength threshold. When the predicted strength is lower than the minimum threshold, it means that there is crack development between the two change points. When the strength is between the maximum and minimum thresholds, it means that there is a weak interlayer between the two change points. When the strength is higher than the maximum strength threshold, it means that there is a roof rock layer between the two change points.

5. The method for identifying the risk of roadway roof instability based on detection while drilling technology according to claim 1, characterized in that: In step S4, The calculation equation for the range of the top plate loose circle is as follows: Where: a is the roadway radius, m is the lithology-related constant, σ c is the uniaxial compressive strength of the immediate roof, s is the constant related to the degree of roof crushing, P0 is the original vertical stress of the roadway, and R0 is the range of the loose zone of the roof; The calculation equation for the height of the self-stabilizing balance arch of the roof is as follows: Where: f is the immediate roof rock strength coefficient, λ is the lateral pressure coefficient, 2a is the tunnel width, f c is the mining impact coefficient; The calculation equation for the roof separation at the weak interlayer position is as follows: Where: η i =√P i / E i I i is the horizontal force at the end of the i-th layer rock beam; E i is the elastic modulus of the i-th rock beam; I i is the cross-sectional inertia moment of the i-th layer of rock beam; q i is the vertical load on the i-th layer of rock beam, γ i is the bulk density of the i-th layer of rock beam; h i is the thickness of the i-th rock beam, w i, is the subsidence of the i-th layer of rock beam, l i is the span of the i-th rock beam, p i Horizontal force at the fixed end of the i-th layer rock beam, η i =p i / E i I i ; The calculation equation of the direct top stress intensity index is as follows: Where: k d is the balance coefficient, σ c is the uniaxial compressive strength of the direct top, h d is the thickness of the direct roof, σ H is the vertical principal stress of the tunnel surrounding rock, R c It is the strength of the immediate top rock layer.

6. The method for identifying the risk of roadway roof instability based on detection while drilling technology according to claim 1, characterized in that: The specific method of step S5 is: S51: The roadway roof instability risk level is divided into five levels: safe T1, basically safe T2, critical safe T3, basically dangerous T4 and dangerous T5; S52: The power set within the comprehensive assessment framework for instability risk contains 32 elements: S53: For the recognition framework Θ, the quality function of any subset A is its power set 2 Θ The mapping to [0,1] is represented by m:2 Θ →[0,1], m is the quality function of Θ on the recognition frame, and its calculation formula is as follows: Where: x is the characteristic value of the evaluation index, Identify the center of the interval for each evaluation indicator, and are the upper and lower limits of each identification interval respectively; when When , the membership degree reaches the maximum value of 1, indicating a state of complete trust; S54: Definition The trust function is Bel:2 Θ →[0,1], Bel(A) represents the total trust level of the evidence on subset A, which constitutes the lower bound of the probability distribution of subset A; the trust function satisfies the following conditions: S55: Definition The likelihood function is Pl:2 Θ →[0,1], Pl(A) constitutes the upper limit of the probability distribution of subset A; the likelihood function satisfies the following conditions: S56: Assume m1, m2, ..., m N Corresponding to A1, A2, ..., A N The quality function of , then the synthesis formula of N pieces of evidence information is as follows: Where: represents the orthogonal sum, K is the evidence conflict coefficient, the larger the K is, the greater the conflict between the two evidences is, and the condition for the two evidence indicators to be fused through the Dempster combination rule is 1 <K; S57: Substitute the instability risk evaluation index value into the DS evidence theory evaluation model to obtain the roof instability risk probability of each drilling location, and draw the instability risk zoning cloud map of different areas of the tunnel roof.

Images