A method for early warning of underground chamber surrounding rock disaster probability based on thermal infrared monitoring

Abstract

The present invention relates to the field of geological exploration technology, and discloses a method for early warning of the probability of underground chamber surrounding rock disaster based on thermal infrared monitoring, comprising the following steps: step S1, using a hydraulic servo press to carry out indoor uniaxial or multiaxial compression tests of surrounding rock samples; step S2, using an infrared thermal imager to collect infrared temperature field data and digital image grayscale data of the sample during the load deformation process; step S3, statistically analyzing the infrared temperature field data and digital image grayscale data of the rock surface, and calculating the infrared extreme value data and grayscale mean value data of the rock surface. The early warning method proposed by the present invention is relatively simple, does not require the use of a large number of sensors, and can obtain multivariate monitoring information of rocks only by infrared thermal imagers, so that timely prediction and early warning can be carried out even in the early stages of the project.

Description

Technical Field

[0001] The present invention relates to the field of underground exploration technology, and in particular to a method for early warning of the probability of underground chamber surrounding rock disaster based on thermal infrared monitoring. Background Art

[0002] Rock instability and failure are the main causes of geological disasters in mines and tunnel engineering. The prediction and early warning of rock instability have always been the core issues faced by the field of rock engineering. In rock engineering, rock instability usually exhibits characteristics such as suddenness, spatiotemporal uncertainty, and difficulty in accurate prediction. Therefore, it has always been a challenging task to establish real-time and efficient observation methods, construct corresponding analysis methods, and build early warning systems. At present, the prediction and early warning methods for rock disasters are mostly based on the monitoring data of rocks during the load and deformation process, analyzing their evolution laws and then determining the precursor information of instability. Most of the precursor indicators are single indicators, and they fail to fully exploit and utilize multivariate monitoring information. In addition, there is a lack of quantitative basis for the determination of the early warning threshold, and there is also a lack of theoretical basis for the joint early warning of multivariate indicators. Therefore, it is of great significance to explore a joint probability early warning method of multivariate indicators that is convenient for engineering application. Summary of the invention

[0003] In order to solve the technical problems raised in the background technology, the present invention provides a method for early warning of the probability of disasters in surrounding rocks of underground chambers based on thermal infrared monitoring.

[0004] The present invention is implemented by the following technical solution: a method for early warning of the probability of underground chamber surrounding rock disaster based on thermal infrared monitoring, comprising the following steps:

[0005] Step S1, using a hydraulic servo press to carry out an indoor uniaxial or multiaxial compression test of surrounding rock samples;

[0006] Step S2, using an infrared thermal imager to collect infrared temperature field data and digital image grayscale data of the sample during the load deformation process;

[0007] Step S3, performing statistical analysis on the infrared temperature field data and digital image grayscale data of the rock surface, and calculating the infrared extreme value data and grayscale mean value data of the rock surface;

[0008] Step S4, using a time sliding window data analysis method to segment the recorded data, and calculate the cosine similarity of the data in each time window;

[0009] Step S5, again use the time sliding window to segment the temperature extreme difference (infrared extreme difference data) and the grayscale mean cosine similarity data, and set the initial threshold, calculate the probability of the data in each time window exceeding the threshold, and name them P(A) and P(B) respectively;

[0010] Step S6: Calculate the posterior probability P(B|A) by the Bayesian formula, and judge whether the rock is about to become unstable and damaged based on the change trend of the posterior probability P(B|A). The present invention provides a simple and easy-to-operate new method for predicting and warning of rock deformation, instability and damage.

[0011] As a further improvement of the above solution, the compression test in step S1 is stopped until the sample is destroyed.

[0012] As a further improvement of the above solution, step S3 includes:

[0013] Step S301: The infrared temperature field data and digital image grayscale data of the rock surface area during the experiment are intercepted by video processing, and the two types of data are statistically analyzed respectively to calculate the infrared extreme value data and grayscale mean value data of the rock surface.

[0014] Step S302: Align the infrared range data and grayscale mean data of the rock surface to ensure the consistency of the data. Figure 2 shown.

[0015] As a further improvement of the above solution, step 4 is specifically as follows:

[0016] The sliding window data analysis method is used to segment the infrared extreme value data and grayscale mean data, and the cosine similarity of the data in adjacent time windows is calculated, such as Figure 3 As shown, the calculation formula of cosine similarity is as follows:

[0017]

[0018] As a further improvement of the above solution, step 5 includes:

[0019] Step S501: Use the data analysis method of segmentation window to segment the cosine similarity of each data, and set initial thresholds respectively; when data in a group of data exceeds the set threshold, the group of data is marked and a new data set is established.

[0020] Step S502: The new data set is again divided into windows, and the proportion of the marked data in each window is calculated to obtain the probability that the infrared extreme value data and the grayscale mean value data exceed the threshold, which are recorded as P(A) and P(B) respectively.

[0021] As a further improvement of the above solution, step 6 is specifically as follows:

[0022] Step S601: In the window segmented in step S402, the proportion of the data in each window that is marked with both the infrared extreme value data and the grayscale mean data is calculated, and the probability that the infrared extreme value data and the grayscale mean data exceed the threshold at the same time can be obtained, which is recorded as P(AB). The prior probability P(A|B) can be obtained by calculating P(AB) / P(B).

[0023] Step S602: The posterior probability P(B|A) can be calculated by the Bayesian formula. The Bayesian formula is as follows:

[0024]

[0025] When the posterior probability reaches 1, it can be considered that the rock is about to be destroyed, such as Figure 4 As shown in Figure 2, the time when the posterior probability reaches 1 is 41 seconds earlier than the rock failure time.

[0026] Compared with the prior art, the present invention has the following beneficial effects:

[0027] The early warning method proposed in the present invention uses the Bayesian principle to couple multivariate precursor indicators to monitor and warn surrounding rock disasters, fully exploits multivariate monitoring information, and determines the quantitative early warning threshold through the size of the posterior probability, which can achieve early warning and ensure the safety of personnel and equipment. The early warning method proposed in the present invention is relatively simple, does not require the use of a large number of sensors, and can obtain multivariate monitoring information of rocks only by infrared thermal imagers, so that timely prediction and early warning can be carried out even in the early stages of the project. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Figure 1 A flowchart of the early warning method proposed by the present invention;

[0029] Figure 2 The parameter evolution diagram of the present invention for aligning infrared extreme value data and grayscale mean value data;

[0030] Figure 3 It is the cosine similarity graph of the infrared extreme difference data and grayscale mean data proposed by the present invention;

[0031] Figure 4 This is the evolution diagram of the Bayesian parameters proposed in the present invention. DETAILED DESCRIPTION

[0032] The present invention is further described below in conjunction with the accompanying drawings and specific implementation methods. It should be noted that, under the premise of no conflict, the various embodiments or technical features described below can be arbitrarily combined to form a new embodiment.

[0033] Embodiment 1:

[0034] A method for early warning of underground chamber surrounding rock disaster probability based on thermal infrared monitoring, comprising the following steps:

[0035] Step S1, using a hydraulic servo press to carry out an indoor uniaxial or multiaxial compression test of surrounding rock samples;

[0036] Step S2, using an infrared thermal imager to collect infrared temperature field data and digital image grayscale data of the sample during the load deformation process;

[0037] Step S3, performing statistical analysis on the infrared temperature field data and digital image grayscale data of the rock surface, and calculating the infrared extreme value data and grayscale mean value data of the rock surface;

[0038] Step S4, using a time sliding window data analysis method to segment the recorded data, and calculate the cosine similarity of the data in each time window;

[0039] Step S5, again use the time sliding window to segment the temperature extreme difference (infrared extreme difference data) and the grayscale mean cosine similarity data, and set the initial threshold, calculate the probability of the data in each time window exceeding the threshold, and name them P(A) and P(B) respectively;

[0040] Step S6: Calculate the posterior probability P(B|A) by the Bayesian formula, and judge whether the rock is about to become unstable and damaged based on the change trend of the posterior probability P(B|A). The present invention provides a simple and easy-to-operate new method for predicting and warning of rock deformation, instability and damage.

[0041] As a further improvement of the above solution, the compression test in step S1 is stopped until the sample is destroyed.

[0042] As a further improvement of the above solution, step S3 includes:

[0043] Step S301: The infrared temperature field data and digital image grayscale data of the rock surface area during the experiment are intercepted by video processing, and the two types of data are statistically analyzed respectively to calculate the infrared extreme value data and grayscale mean value data of the rock surface.

[0044] Step S302: Align the infrared range data and grayscale mean data of the rock surface to ensure the consistency of the data. Figure 2 shown.

[0045] As a further improvement of the above solution, step 4 is specifically as follows:

[0046] The sliding window data analysis method is used to segment the infrared extreme value data and grayscale mean data, and the cosine similarity of the data in adjacent time windows is calculated, such as Figure 3 As shown, the calculation formula of cosine similarity is as follows:

[0047]

[0048] As a further improvement of the above solution, step 5 includes:

[0049] Step S501: Use the data analysis method of segmentation window to segment the cosine similarity of each data, and set initial thresholds respectively; when data in a group of data exceeds the set threshold, the group of data is marked and a new data set is established.

[0050] Step S502: The new data set is again divided into windows, and the proportion of the marked data in each window is calculated to obtain the probability that the infrared extreme value data and the grayscale mean value data exceed the threshold, which are recorded as P(A) and P(B) respectively.

[0051] As a further improvement of the above solution, step 6 is specifically as follows:

[0052] Step S601: In the window segmented in step S402, the proportion of the data in each window that is marked with both the infrared extreme value data and the grayscale mean data is calculated, and the probability that the infrared extreme value data and the grayscale mean data exceed the threshold at the same time can be obtained, which is recorded as P(AB). The prior probability P(A|B) can be obtained by calculating P(AB) / P(B).

[0053] Step S602: The posterior probability P(B|A) can be calculated by the Bayesian formula. The Bayesian formula is as follows:

[0054]

[0055] When the posterior probability reaches 1, it can be considered that the rock is about to be destroyed, such as Figure 4 As shown in Figure 2, the time when the posterior probability reaches 1 is 41 seconds earlier than the rock failure time.

[0056] The above-mentioned embodiments are only preferred embodiments of the present invention and cannot be used to limit the scope of protection of the present invention. Any non-substantial changes and substitutions made by technicians in this field on the basis of the present invention shall fall within the scope of protection required by the present invention.

Claims

1. A method for early warning of the probability of underground chamber surrounding rock disaster based on thermal infrared monitoring, characterized in that: The following steps are involved: Step S1, using a hydraulic servo press to carry out an indoor uniaxial or multiaxial compression test of surrounding rock samples; Step S2, using an infrared thermal imager to collect infrared temperature field data and digital image grayscale data of the sample during the load deformation process; Step S3, performing statistical analysis on the infrared temperature field data and digital image grayscale data of the rock surface, and calculating the infrared extreme value data and grayscale mean value data of the rock surface; Step S4, using a time sliding window data analysis method to segment the recorded data, and calculate the cosine similarity of the data in each time window; Step S5, use the time sliding window again to segment the data of temperature extreme difference and grayscale mean cosine similarity, set the initial threshold, calculate the probability of data exceeding the threshold in each time window, and name them P(A) and P(B) respectively; Step S6, calculating the posterior probability P(B|A) by using the Bayesian formula, and judging whether the rock is about to become unstable and fail based on the change trend of the posterior probability P(B|A); The step 4 is specifically as follows: The sliding window data analysis method is used to segment the infrared extreme value data and grayscale mean value data, and the cosine similarity of the data in adjacent time windows is calculated. The calculation formula of the cosine similarity is as follows: The step 5 comprises: Step S501: segment the cosine similarity of each data using the data analysis method of the segmented window, and set initial thresholds respectively; when data in a group of data exceeds the set threshold, mark the group of data and create a new data set; Step S502: The new data set is again divided into windows, and the proportion of the marked data in each window is calculated, so as to obtain the probability of the infrared extreme value data and the gray mean value data exceeding the threshold, which are recorded as P(A) and P(B) respectively; The step 6 is specifically as follows: Step S601: In the window segmented in step S402, the proportion of the data marked with both the infrared extreme value data and the grayscale mean data in each window is calculated, so as to obtain the probability that both the infrared extreme value data and the grayscale mean data exceed the threshold at the same time, which is recorded as P(AB). The prior probability P(A|B) can be obtained by calculating P(AB) / P(B); Step S602: The posterior probability P(B|A) can be calculated by the Bayesian formula, which is as follows: When the posterior probability reaches 1, it can be considered that the rock is about to fail.

2. The method for early warning of the probability of underground chamber surrounding rock disaster based on thermal infrared monitoring as claimed in claim 1, characterized in that: The compression test in step S1 is stopped until the sample is destroyed.

3. The method for early warning of the probability of underground chamber surrounding rock disaster based on thermal infrared monitoring as claimed in claim 1, characterized in that: The step S3 comprises: Step S301: intercepting infrared temperature field data and digital image grayscale data of the rock surface area during the experiment through video processing, and performing statistical analysis on the two types of data respectively, and calculating infrared extreme value data and grayscale mean value data of the rock surface; Step S302: Align the infrared range data and grayscale mean data of the rock surface to ensure the consistency of the data volume.

Images