Rock burst risk assessment method based on confidence ellipse

Abstract

The invention discloses a rock burst risk assessment method based on a confidence ellipse, and relates to the field of rock burst risk assessment under a mine, and the method specifically comprises the following steps: S1, collecting the energy data in the mine in real time through a sensor, and recording the propulsion degree data related to mining; s2, preprocessing the collected energy data; s3, determining a most suitable probability distribution model; s4, calculating the probability of the energy data; and S5, introducing propulsion degree data, setting three confidence levels of 90%, 95% and 99%, constructing a confidence ellipse, and evaluating the rock burst risk according to the position of the energy data in the confidence ellipse and the probability of the corresponding energy data. According to the method, the preprocessed energy data is fitted by using multiple probability distribution models, and the most suitable probability distribution model is determined according to the goodness of fit, so that the adaptability of rock burst risk assessment to different engineering environments and the assessment accuracy are improved.

Description

Technical Field

[0001] The invention relates to the field of rock burst hazard assessment in mines, and in particular to a rock burst hazard assessment method based on a confidence ellipse. Background Art

[0002] In the process of evaluating the rock burst hazard of rock burst mines, accurate assessment methods are crucial to ensure mine safety. Energy data, such as microseismic energy data or ground sound energy data, is a key factor in evaluating the rock burst hazard, which contains important information about potential impact risks.

[0003] Traditional rock burst hazard assessment is usually based on empirical formulas and fixed parameters. Specific empirical formulas are summarized through historical data and field observations, and some measurable parameters are substituted into the formula to calculate the rock burst hazard index. However, the empirical formulas and fixed parameters in this type of method are determined based on specific engineering environments and historical data. When faced with different engineering environments, these fixed parameters and formulas cannot accurately reflect the actual situation, resulting in a large deviation between the evaluation results and the actual rock burst hazard. At the same time, some studies have tried to fit energy data using normal distribution or skewed distribution, and evaluate rock burst hazard through analysis of a large amount of energy data and abnormal coefficients. For example, when the data point deviates from the mean of the distribution by more than a certain standard deviation range, it may indicate a higher rock burst hazard. However, these methods often assume that the energy data completely conforms to a certain specific distribution, which is difficult to adapt to changes in different engineering environments, and the evaluation results also have large deviations. Summary of the invention

[0004] The present invention provides a rock burst hazard assessment method based on confidence ellipse to solve the problem that the existing assessment method is difficult to adapt to the changes in different engineering environments and the assessment results have large deviations.

[0005] The present invention provides a rock burst hazard assessment method based on a confidence ellipse, which adopts the following technical solution: the method comprises the following steps:

[0006] S1. Use sensors to collect energy data in the mine in real time and record the advancement data related to mining;

[0007] S2, preprocessing the collected energy data;

[0008] S3, using a variety of probability distribution models to fit the preprocessed energy data, and then comparing the goodness of fit under different probability distribution models, so as to determine the most suitable probability distribution model;

[0009] S4. Calculate the probability of energy data according to the determined probability distribution model;

[0010] S5. Based on the probability of energy data, the advancement data is introduced, and three confidence levels of 90%, 95%, and 99% are set. A confidence ellipse is constructed with the horizontal and vertical coordinates being the advancement and energy data values, respectively. The rock burst hazard is evaluated based on the position of the energy data within the confidence ellipse. The specific evaluation rules are as follows:

[0011] When the energy data is within the confidence ellipse of the 90% confidence level, it is judged that there is no rock burst hazard;

[0012] When the energy data is outside the confidence ellipse of the 90% confidence level but inside the confidence ellipse of the 95% confidence level, it is judged as a low rock burst hazard;

[0013] When the energy data is outside the confidence ellipse of the 95% confidence level but inside the confidence ellipse of the 99% confidence level, it is judged as a medium rock burst hazard;

[0014] When the energy data is outside the confidence ellipse at the 99% confidence level, it is judged to be a high rock burst hazard.

[0015] Preferably, in step S2, preprocessing the collected energy data includes removing erroneous data and outliers and performing normalization processing.

[0016] Preferably, in step S3, the multiple probability distribution models include normal distribution, Weibull distribution, generalized extreme value distribution and tlocation-scale distribution.

[0017] Preferably, in step S5, the specific method of constructing the confidence ellipse is:

[0018] The mean and covariance matrix of the energy data are calculated by statistical methods, and then the confidence ellipse is constructed using the mean and covariance matrix. The equation of the confidence ellipse is:

[0019] (x-μ) T Σ -1 (x-μ)=c

[0020] In the above equation, x is the energy data vector, μ is the mean vector of the energy data, Σ is the covariance matrix, and c is a constant related to the confidence level, which is determined by looking up the chi-square distribution table.

[0021] Compared with the prior art, the present invention has the following beneficial effects:

[0022] 1. The present invention takes into account the differences in different engineering environments, uses a variety of probability distribution models to fit the preprocessed energy data respectively, and then compares the goodness of fit under different probability distribution models to determine the most suitable probability distribution model, thereby improving the adaptability of rock burst hazard assessment to different engineering environments, and also improving the accuracy of rock burst hazard assessment.

[0023] 2. The present invention comprehensively considers the characteristics of the confidence ellipse of energy data at different confidence levels to avoid misjudgment of rock burst hazard and achieve a more comprehensive rock burst hazard assessment.

[0024] 3. The present invention constructs a confidence ellipse whose horizontal and vertical coordinates are respectively the advancement degree and energy data values, so that the evaluation result can comprehensively consider the change of the advancement degree, further improving the accuracy of the evaluation. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0026] Figure 1 is a flow chart of the present invention;

[0027] Figure 2 It is a probability density distribution fitting diagram of microseismic energy under various probability distribution models of the present invention;

[0028] Figure 3 It is a probability density distribution fitting diagram of microseismic energy in different confidence intervals under the t location-scale probability distribution model of the present invention;

[0029] Figure 4 This is a diagram of the rock burst hazard assessment results of the present invention with the introduction of advancement. DETAILED DESCRIPTION

[0030] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0031] An embodiment of a rock burst hazard assessment method based on a confidence ellipse of the present invention is as follows: Figure 1 As shown, the method comprises the following steps:

[0032] S1. Use sensors to collect energy data in the mine in real time and record the advancement data related to mining;

[0033] S2, preprocessing the collected energy data;

[0034] S3, using a variety of probability distribution models to fit the preprocessed energy data, and then comparing the goodness of fit under different probability distribution models, so as to determine the most suitable probability distribution model;

[0035] S4. Calculate the probability of energy data according to the determined probability distribution model;

[0036] S5. Based on the probability of energy data, the advancement data is introduced, and three confidence levels of 90%, 95%, and 99% are set. A confidence ellipse is constructed with the horizontal and vertical coordinates being the advancement and energy data values, respectively. The rock burst hazard is evaluated based on the position of the energy data within the confidence ellipse. The specific evaluation rules are as follows:

[0037] When the energy data is within the confidence ellipse of the 90% confidence level, it is judged that there is no rock burst hazard;

[0038] When the energy data is outside the confidence ellipse of the 90% confidence level but inside the confidence ellipse of the 95% confidence level, it is judged as a low rock burst hazard;

[0039] When the energy data is outside the confidence ellipse of the 95% confidence level but inside the confidence ellipse of the 99% confidence level, it is judged as a medium rock burst hazard;

[0040] When the energy data is outside the confidence ellipse at the 99% confidence level, it is judged to be a high rock burst hazard.

[0041] In the present embodiment, it should be noted that, in step S1, microseismic sensors and borehole stress sensors are arranged at different key locations of the rock burst mine, such as coal mining faces, tunnels and other areas prone to rock burst, to ensure that the two sensors can accurately monitor the changes in physical quantities related to the danger of rock burst. The two sensors collect energy data in real time and transmit the energy data to the monitoring system for storage through wired or wireless data transmission devices, thereby obtaining dynamic indicator parameter data with time series characteristics, such as data on the changes in microseismic energy over time.

[0042] In this embodiment, it should be noted that in step S2, accurate and pre-processed energy data is the basis for subsequent analysis, which can improve the reliability of the analysis results. The data collected by sensors at different locations can fully reflect the stress and energy changes in the mine.

[0043] In this embodiment, it should be noted that Figure 2 As shown, it is a probability density distribution fitting diagram of microseismic energy under various probability distribution models. Since the probability is obtained by integrating the probability density over the corresponding interval, if the probability density is high, then the probability obtained by integrating over the interval is relatively large. Therefore, the probability density distribution fitting diagram of microseismic energy under various probability distribution models can intuitively reflect the probability distribution fitting of microseismic energy under various probability distribution models. In step S3, different probability distribution models have different characteristics. Through fitting and comparison of multiple models, the complex probability characteristics of energy data can be better captured, which provides a basis for accurately constructing confidence ellipses and evaluating the hazard of rock burst. The attempt of multiple distribution models can adapt to data distributions of different shapes and avoid the limitations of a single model. The goodness of fit index can objectively evaluate the degree of fit between the model and the data, thereby selecting the model that best represents the data distribution.

[0044] In this embodiment, it should be noted that Figure 3 As shown, it is a probability density distribution fitting diagram of microseismic energy in different confidence intervals under the t location-scale probability distribution model, which can also intuitively reflect the probability distribution fitting of microseismic energy in different confidence intervals under the t location-scale probability distribution model. In step S4, the probability of energy data is calculated according to the determined probability distribution model. Specifically, the probability of energy data in different intervals is calculated according to the determined probability distribution model. Calculating the probability of different intervals can further understand the possibility of energy data appearing in various ranges, and provide a basis for subsequent hazard judgment. For example, when it is determined that the energy data conforms to the normal distribution and its mean and standard deviation are known, the probability of energy data in a certain specific interval (such as [E1, E2]) can be calculated. The calculation formula is:

[0045]

[0046] Among them, μ is the mean, σ is the standard deviation, and X is the energy data variable. The integral can be solved by numerical calculation methods (such as trapezoidal method, Simpson method, etc.) to obtain the specific probability value.

[0047] In this embodiment, it should be noted that Figure 4As shown in the figure, it is a rock burst hazard assessment result diagram with the introduction of advancement. In step S5, the advancement is combined with energy data to construct a confidence ellipse based on the following principle: in rock burst mines, the rock burst hazard is not only related to the size and distribution of energy, but also closely related to the advancement during the mining process. The change of advancement will affect the distribution of the stress field, and then affect the accumulation and release of energy. For example, in the early stage of mining, when the advancement is slow, the stress adjustment is relatively stable, and the energy accumulation rate is slow. As the advancement increases, it may lead to increased stress concentration, accelerated energy accumulation, and a corresponding increase in rock burst hazard. Therefore, taking the advancement as a dimension and considering it in combination with energy data can more comprehensively reflect the dynamic changes of rock burst hazard.

[0048] In this embodiment, it should be noted that in step S5, if the energy data is within the confidence ellipse of the 90% confidence level, the probability of the energy data in this area is high (the probability within the confidence ellipse is high), and it is determined that there is no rock burst risk. This is because the current mining conditions are in a relatively stable range, the fluctuation of energy data is small and closer to the normal range, and the possibility of rock burst is low, so it is determined that there is no rock burst risk. For example, when analyzing microseismic energy data, if the energy value at a certain moment is within the 99% confidence ellipse, it means that the energy state at this time is relatively stable and the probability of rock burst is small.

[0049] If the energy data is outside the confidence ellipse of the 90% confidence level but inside the confidence ellipse of the 95% confidence level, it is judged as a low rock burst hazard. This is because the energy data-advancement combination deviates from the relatively stable core area, and its probability of occurrence is relatively reduced. At this time, the data points are in a relatively unstable area, the fluctuation range of the energy data has increased, and there is a certain possibility of rock burst hazard. In this case, mining operations can continue, but the frequency and accuracy of monitoring need to be appropriately increased, and the changing trends of energy data and advancement data need to be closely monitored, especially factors that may lead to intensified energy fluctuations, such as slight changes in geological conditions and the operating status of mining equipment. At the same time, preventive support inspections and reinforcements can be carried out in local areas where potential risks may exist to enhance the stability of the mine structure and ensure safe mining under weak rock burst hazard conditions.

[0050] If the energy data is outside the confidence ellipse of the 95% confidence level but inside the confidence ellipse of the 99% confidence level, it is judged as a medium rock burst hazard. At this time, the energy-advancement combination has obviously deviated from the normal stable state, and its probability of occurrence is further reduced, which indicates that the possibility of rock burst has increased significantly, reaching the medium rock burst hazard level. At this time, a series of positive and effective measures must be taken immediately to deal with it. First, the mining operation parameters should be decisively adjusted, such as appropriately reducing the mining speed, optimizing the mining sequence, and avoiding excessive concentration of mining operations in high-risk areas to reduce the speed and degree of energy accumulation.

[0051] If the energy data is outside the confidence ellipse of the 99% confidence level, it indicates that the energy-advance combination is in an extremely rare and extremely unstable state, and is judged to be a high rock burst risk. Once the microseismic energy data is monitored to be in this state, emergency measures should be taken immediately, such as stopping mining operations, increasing support strength on a large scale, and implementing emergency pressure relief measures to ensure that the situation can be effectively controlled in the strong impact danger state to prevent rock burst accidents.

[0052] In this embodiment, in step S2, preprocessing the collected energy data includes removing erroneous data and abnormal values ​​and performing normalization processing.

[0053] In this embodiment, it should be noted that removing erroneous data and outliers can remove the interference of erroneous data and outliers on the analysis results, making it easier to perform operations such as analysis of various probability distribution models.

[0054] In this embodiment, in step S3, the multiple probability distribution models include normal distribution, Weibull distribution, generalized extreme value distribution and t location-scale distribution.

[0055] In this embodiment, it should be noted that for different confidence levels, the boundaries of the confidence interval are obtained by calculation according to the corresponding statistical theory and the characteristics of the distribution model. For example, for a normal distribution, the boundaries of a 95% confidence interval are μ±1.96σ, where μ is the mean and σ is the standard deviation; for other distributions, they can be obtained by consulting relevant statistical tables or using specific statistical software; for example, for a Weibull distribution, the corresponding confidence interval boundary values ​​can be calculated according to its distribution function and confidence level.

[0056] In this embodiment, it should be noted that for each probability distribution model, the model parameters are calculated using corresponding statistical methods to make the model as close to the actual data distribution as possible. For example, for the normal distribution, its parameters are determined by calculating the sample mean and sample standard deviation. For the Weibull distribution, its shape parameters and scale parameters can be determined using methods such as the maximum likelihood estimation method. The most suitable probability distribution model is determined by comparing the goodness of fit of different probability distribution models. The comparison of goodness of fit can use indicators such as root mean square error and determination coefficient. The smaller the root mean square error, the closer the model prediction value is to the actual value, and the closer the determination coefficient is to 1, the better the model fits the data.

[0057] In this embodiment, in step S5, the specific method of constructing the confidence ellipse is:

[0058] The mean and covariance matrix of the energy data are calculated by statistical methods, and then the confidence ellipse is constructed using the mean and covariance matrix. The equation of the confidence ellipse is:

[0059] (x-μ) T Σ -1 (x-μ)=c

[0060] In the above equation, x is the energy data vector, μ is the mean vector of the energy data, Σ is the covariance matrix, and c is a constant related to the confidence level, which is determined by looking up the chi-square distribution table.

[0061] In this embodiment, it should be noted that, for example, for a 95% confidence level, the value of c can be determined according to a chi-square distribution with a degree of freedom of 2. In the specific calculation, firstly, the propulsion mean and energy mean, as well as each element σ of the covariance matrix Σ are calculated according to the collected propulsion and energy data. ij (i, j = 1, 2), and then look up the chi-square distribution table according to the selected confidence level to get the c value, thereby determining the equation of the confidence ellipse.

[0062] The above are only preferred embodiments of the present invention and are 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 protection scope of the present invention.

Claims

1. A rock burst hazard assessment method based on confidence ellipse, characterized in that: The following steps are involved: S1. Use sensors to collect energy data in the mine in real time and record the advancement data related to mining; S2, preprocessing the collected energy data; S3, using a variety of probability distribution models to fit the preprocessed energy data, and then comparing the goodness of fit under different probability distribution models, so as to determine the most suitable probability distribution model; S4. Calculate the probability of energy data according to the determined probability distribution model; S5. Based on the probability of energy data, the advancement data is introduced, and three confidence levels of 90%, 95%, and 99% are set. A confidence ellipse is constructed with the horizontal and vertical coordinates being the advancement and energy data values, respectively. The rock burst hazard is evaluated based on the position of the energy data within the confidence ellipse. The specific evaluation rules are as follows: When the energy data is within the confidence ellipse of the 90% confidence level, it is judged that there is no rock burst hazard; When the energy data is outside the confidence ellipse of the 90% confidence level but inside the confidence ellipse of the 95% confidence level, it is judged as a low rock burst hazard; When the energy data is outside the confidence ellipse of the 95% confidence level but inside the confidence ellipse of the 99% confidence level, it is judged as a medium rock burst hazard; When the energy data is outside the confidence ellipse at the 99% confidence level, it is judged to be a high rock burst hazard.

2. The method for assessing rock burst risk based on confidence ellipse according to claim 1, characterized in that: In step S2, the collected energy data is preprocessed, including removing erroneous data and outliers and performing normalization processing.

3. The method for assessing rock burst risk based on confidence ellipse according to claim 1, characterized in that: In step S3, the multiple probability distribution models include normal distribution, Weibull distribution, generalized extreme value distribution and tlocation-scale distribution.

4. The method for assessing rock burst risk based on confidence ellipse according to claim 1, characterized in that: In step S5, the specific method of constructing the confidence ellipse is: The mean and covariance matrix of the energy data are calculated by statistical methods, and then the confidence ellipse is constructed using the mean and covariance matrix. The equation of the confidence ellipse is: (x-μ) T S -1 (x-μ)=c In the above equation, x is the energy data vector, μ is the mean vector of the energy data, Σ is the covariance matrix, and c is a constant related to the confidence level, which is determined by looking up the chi-square distribution table.

Images