A single-axis compression rock instability prediction and early warning method based on multi-source data fusion
By using multi-source data fusion technology, rock sample data can be acquired and quantified in real time. Combined with DS evidence theory, rock instability prediction and early warning can be carried out, which solves the problem that single index data is difficult to fully characterize rock instability and fracture, and improves the accuracy of prediction and the reliability of risk assessment.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- NORTHEASTERN UNIV CHINA
- Filing Date
- 2023-10-23
- Publication Date
- 2026-07-28
AI Technical Summary
In existing technologies, rock instability prediction and early warning methods in mining engineering mostly rely on single-index data, which makes it difficult to comprehensively characterize the rock instability and fracture process, thus affecting the accuracy of prediction.
A multi-source data fusion method is adopted, which uses devices such as DHDAS, PCI-2, and EOS 90D to acquire strain, acoustic emission, and speckle data in real time. Data quantification is performed using LabVIEW secondary development and optical flow method, and data fusion is combined with DS evidence theory to finally give an early warning of damage time and risk level.
It enables multi-dimensional characterization of rock instability processes, improves the accuracy of prediction and early warning and the reliability of risk assessment, and provides early warning results earlier and more accurately.
Smart Images
Figure CN117451499B_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of experimental data information technology in rock mechanics experiments, and more particularly to a method for predicting and warning of uniaxial compression rock instability based on multi-source data fusion. Background Technology
[0002] Mining sites present complex conditions, with numerous unstable and uncontrollable factors affecting mine safety in addition to mechanistic factors. This hinders the exploration of rock instability and failure mechanisms. Furthermore, the limitations of mining conditions restrict field testing, impacting the innovation and development of theories, methods, and technologies. Therefore, conducting research on rock damage and instability characteristics through indoor rock mechanics experiments will provide a crucial theoretical foundation for guiding the prediction and early warning of mine disasters.
[0003] Indoor experiments provide the conditions for multi-source monitoring, enabling the acquisition of multi-source monitoring data during the loading process. This data can reflect the rock instability and failure process from multiple perspectives, including acoustic, optical, and mechanical aspects. Current evaluation methods for predicting and warning of rock instability mostly rely on single-index data for early warning, making it difficult to grasp the full picture of the instability and failure process. Considering multi-source data fusion processing, it is possible to characterize the rock instability and fracture process from multiple dimensions, more closely reflecting the actual instability and fracture situation. Summary of the Invention
[0004] To address the aforementioned technical problems, this invention provides a method for predicting and warning of uniaxial compression rock instability based on multi-source data fusion. This method is designed for uniaxial compression experiments conducted in a laboratory setting. Monitoring data is acquired in real-time using monitoring equipment. The damage to the rock sample is quantified based on the characteristics of the monitoring data. The DS evidence theory is introduced to fuse the quantified damage data, ultimately predicting the failure time and providing early warning results based on risk classification according to the data fusion results.
[0005] The technical means employed in this invention are as follows: A method for predicting and warning of uniaxial compressive rock instability based on multi-source data fusion includes: S1. Acquire monitoring data of rock samples during uniaxial compression loading and perform damage quantification on the data in real time; S2. Based on the damage quantification results of multi-source monitoring data, the data are fused using the DS evidence theory; S3. Based on the single-index damage quantification results, a prediction of the damage time is given, and an early warning result is given according to the risk classification level based on the data fusion results.
[0006] Further, step S1 includes: S11. The rock sample was monitored during uniaxial compression loading using DHDAS, PCI-2, and EOS 90D. Strain and acoustic emission data were acquired in real time using LabVIEW secondary development, and speckle data were acquired in real time using EOS Unity. S12. For the strain data acquired in real time, define strain damage quantification based on the volume strain inflection point; S13. For the real-time acquired acoustic emission data, define the acoustic emission data quantification damage based on the dimensionless curve of the acoustic emission parameters; S14. For the speckle data acquired in real time, the damage is quantified based on the speckle data volume strain matrix defined by the optical flow method.
[0007] Further, step S12 specifically includes: S121, through axial strain and transverse strain The strain of the rock sample was calculated as follows:
[0008] It is assumed that damage and accumulation in the rock sample begin after the volumetric strain expansion point. Considering the change in Poisson's ratio during loading, it is assumed that the rock sample completely fails when the volumetric strain curve returns to zero, at which point the corresponding Poisson's ratio is 0.5. Therefore, it is considered that the rock damage reaches its maximum value of 1 when the Poisson's ratio reaches 0.5. S122. Based on the rock expansion point and Poisson's ratio, define the quantitative relationship between rock sample damage and volumetric strain:
[0009] In the formula, This represents the volumetric strain value at the expansion point. This represents the volumetric strain value at the current moment. The damage is calculated based on this formula. When the Poisson's ratio exceeds 0.5, it will be greater than 1. In this case, the rock sample is considered to be completely damaged, and this value is set to 1. S123. For dynamically acquired volumetric strain sequences and the volumetric strain value acquired at the current moment. , constructing a time-based As the index endpoint, Dynamic time window for index length The specific formula is as follows:
[0010] S124, For dynamic time windows Determine the local maximum value of the set according to the following criteria. Is it an inflection point? If the value is greater than 0, then the local maximum value point is considered to be the inflection point of the temporary body strain.
[0011] S125, Temporary body strain inflection points are dynamically stored in a set. In the middle, if the current set If not empty, then the volumetric strain value obtained at the current moment is used. To determine failure based on the values in the set of strain inflection points of a temporary body, and to define the set of failure criteria for strain inflection points of a temporary body:
[0012] in, The length of the set after dynamically inserting the strain inflection point of the temporary body. S126. Set the failure tolerance threshold. Merge temporary strain inflection point set With failure discrimination set The final set of judgments for volumetric strain inflection points is obtained as follows: S127. Assume the duration of a suspected volumetric strain inflection point is... Set the inflection point duration to Add new judgment conditions:
[0013] Thus, the duration was not reached. The inflection point will be eliminated; S128. Select the values of the volumetric strain inflection points from the set:
[0014] S129. After calculating the inflection point value, the volumetric strain value is obtained in real time. Inflection point of volumetric strain Calculate the real-time damage of the rock sample: .
[0015] Further, step S13 specifically includes: S131. Defining the impairment of acoustic emission data by the tangent angle of the dimensionless curve, quantizing the tangent angle between 45° and 90°, and mapping it to 0 to 1 to represent the degree of impairment; considering that the larger the angle, the more severe the increase in acoustic emission ringing, exponential quantization is adopted:
[0016] S132. Define the time factor:
[0017] In the formula, Indicates the current moment. This indicates the time when the inflection point of volumetric strain is reached. Indicates the predicted time of destruction; S133, Redefining Acoustic Emission Damage:
[0018] S134. For the dynamically acquired ring count sequence, calculate the cumulative ring count corresponding to the current moment in real time. , constructing a time-based For indexing purposes, Dynamic time window for index length Simultaneously, the cumulative ring count is stored in the cumulative ring count array. In the middle, the specific formula is as follows:
[0019]
[0020] S135, Regarding the current dynamic time window The velocity of the current time window is calculated using a difference method. and timing speed Store to velocity sequence In the middle, the specific formula is as follows:
[0021]
[0022] S136. After calculating the inflection point of the stable body strain from the strain data, the velocity series... Find the average velocity during the "uniform deformation" stage before the inflection point, and calculate the cumulative ringing sequence. Dimensionless transformation is performed, as shown in the following formula:
[0023]
[0024] S137. Regarding the obtained dimensionless time series... and the dimensionless value calculated at the current moment. , constructing a time-based For indexing purposes, Dynamic time window for index length :
[0025] S138, Regarding the current dynamic time window The slope of the time window at the current moment is calculated using a difference method. And calculate the slope arctangent value Ultimately Stored in angle sequence In the middle, the specific formula is as follows:
[0026]
[0027]
[0028] S139. Regarding the calculated angle sequence This is quantified as damage, as shown in the following formula: .
[0029] Further, step S14 specifically includes: S141. After obtaining the volumetric strain matrix, the damage on a surface is considered as the accumulation of damage at all points, and the damage at a point is defined as follows:
[0030] In the formula, This represents the volumetric strain value at the expansion point. This represents the volumetric strain value at the current moment. The damage is calculated based on this formula. When the Poisson's ratio exceeds 0.5, it will be greater than 1. In this case, the rock sample is considered to be completely damaged, and its value is set to 1.
[0031] S142. Define the time factor as:
[0032] In the formula, Indicates the current moment. This indicates the time when the inflection point of volumetric strain is reached. Indicates the predicted time of destruction.
[0033] S143. Correct for damage, assuming that at a point, after the inflection point of volumetric strain, a defect occurs. Secondary local damage, then the damage at that point for:
[0034] S144. For dynamically acquired volumetric strain sequences and the volumetric strain value acquired at the current moment. , constructing a time-based As the index endpoint, Dynamic time window for index length The specific formula is as follows:
[0035] S145, For dynamic time windows Determine the local maximum value of the set according to the following criteria. Is it an inflection point? If the value is greater than 0, then the local maximum value point is the inflection point of the temporary body strain.
[0036] S146, Temporary body strain inflection points are dynamically stored in a set. In the middle, if the current set If not empty, then the volumetric strain value obtained at the current moment is used. To determine failure based on the values in the set of strain inflection points of a temporary body, and to define the set of failure criteria for strain inflection points of a temporary body:
[0037] in, The length of the set after dynamically inserting the strain inflection point of the temporary body; S147. Set the failure tolerance threshold. Realize the set of temporary body strain inflection points With failure discrimination set The merging of these results yields the final set of criteria for determining volumetric strain inflection points:
[0038] S148. Select the values of the volumetric strain inflection points from the set. And through the real-time acquired volumetric strain values Inflection point of volumetric strain Calculate the real-time damage of the rock sample:
[0039] For the The nth point, assuming the nth... The point appeared after the inflection point of the volumetric strain. Secondary local damage, then the damage at that point for:
[0040] S149, for dimensions of The volumetric strain matrix, the first Damage at each point is The current real-time damage of a speckle photograph is: .
[0041] Further, step S2 includes: S21. Define Dempster's composition rule for the DS evidence theory; S22. Modify the Dempster composition rule in response to the Zadeh paradox.
[0042] Further, step S21 specifically includes: S211, Regarding , The two mass functions m 1, m The synthesis rules for Dempster 2 are as follows:
[0043] in, K The normalization constant is ; S212, Regarding , A finite number of mass functions on m 1, m 2,…, m n The synthesis rules for Dempster are as follows:
[0044] in, K The normalization constant is .
[0045] Further, step S22 specifically includes: S221. Regarding the Zadeh paradox, the Dempster composition rule is rewritten as follows:
[0046] S222. Based on the rewritten Dempster composition rules in step S221, we have:
[0047] in, This is the allocation function for the probability of evidence conflict.
[0048] Furthermore, in step S3, the prediction of the destruction time is achieved through the inverse velocity method, specifically including: S31, From speckle pattern to orange alert From the start until the acoustic emission reaches the orange alert level. Finally, this interval is defined as the speckle-acoustic emission early warning interval, and the DS fusion damage results of the speckle-acoustic emission early warning interval constitute the damage time series. :
[0049] In the formula, Indicates the interval length. Indicates from arrive DS fusion damage results over the time period; S32. Based on the damage time series Obtain the inverse velocity sequence :
[0050] S33, Using the least squares method A single fitting is performed on the sequence data to obtain the fitted line, and the line is defined as... x The intersection of the axes is The predicted time of destruction is: .
[0051] Furthermore, the warning results in step S3 are characterized by a four-level warning system: blue, yellow, orange, and red.
[0052] Compared with the prior art, the present invention has the following advantages: The present invention provides a method for predicting and warning of uniaxial compression rock instability based on multi-source data fusion. For uniaxial compression experiments conducted in the laboratory, monitoring data is acquired in real time through monitoring equipment. The damage of the rock sample is quantified according to the characteristics of the monitoring data itself. The DS evidence theory is introduced to fuse the quantified damage data. Finally, the failure time is predicted and a warning result is given according to the risk level based on the data fusion result. Attached Figure Description
[0053] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0054] Figure 1 This is a flowchart of the method of the present invention.
[0055] Figure 2 This is a flowchart of the strain damage quantification calculation of the present invention.
[0056] Figure 3 This is a flowchart of the acoustic emission damage quantification calculation for the present invention.
[0057] Figure 4 This is a flowchart of the speckle damage quantification calculation of the present invention.
[0058] Figure 5 This is a schematic diagram illustrating the calculation of the failure time using the reverse velocity method of this invention. Detailed Implementation
[0059] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.
[0060] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.
[0061] like Figure 1 As shown, this invention provides a method for predicting and warning of uniaxial compressive rock instability based on multi-source data fusion, comprising: S1. Acquire monitoring data of rock samples during uniaxial compression loading and perform damage quantification on the data in real time; S2. Based on the damage quantification results of multi-source monitoring data, the data are fused using the DS evidence theory; S3. Based on the single-index damage quantification results, a prediction of the damage time is given, and an early warning result is given according to the risk classification level based on the data fusion results.
[0062] In a specific implementation, as a preferred embodiment of the present invention, step S1 includes: S11. The rock sample was monitored during uniaxial compression loading using DHDAS, PCI-2, and EOS 90D. Strain and acoustic emission data were acquired in real time using LabVIEW secondary development, and speckle data were acquired in real time using EOS Unity. S12. For the strain data acquired in real time, define strain damage quantification based on the volume strain inflection point; S13. For the real-time acquired acoustic emission data, define the acoustic emission data quantification damage based on the dimensionless curve of the acoustic emission parameters; S14. For the speckle data acquired in real time, the damage is quantified based on the speckle data volume strain matrix defined by the optical flow method.
[0063] In specific implementation, as a preferred embodiment of the present invention, such as Figure 2 As shown, step S12 specifically includes: S121, through axial strain and transverse strain The strain of the rock sample was calculated as follows:
[0064] It is assumed that damage and accumulation in the rock sample begin after the volumetric strain expansion point. Considering the change in Poisson's ratio during loading, it is assumed that the rock sample completely fails when the volumetric strain curve returns to zero, at which point the corresponding Poisson's ratio is 0.5. Therefore, it is considered that the rock damage reaches its maximum value of 1 when the Poisson's ratio reaches 0.5. S122. Based on the rock expansion point and Poisson's ratio, define the quantitative relationship between rock sample damage and volumetric strain:
[0065] In the formula, This represents the volumetric strain value at the expansion point. This represents the volumetric strain value at the current moment. The damage is calculated based on this formula. When the Poisson's ratio exceeds 0.5, it will be greater than 1. In this case, the rock sample is considered to be completely damaged, and this value is set to 1. S123. For dynamically acquired volumetric strain sequences and the volumetric strain value acquired at the current moment. , constructing a time-based As the index endpoint, Dynamic time window for index length The specific formula is as follows:
[0066] S124, For dynamic time windows Determine the local maximum value of the set according to the following criteria. Is it an inflection point (i.e., is the curve convex)? If the value is greater than 0, then the local maximum value point is considered to be the inflection point of the temporary body strain.
[0067] S125, Temporary body strain inflection points are dynamically stored in a set. In the middle, if the current set If not empty, then the volumetric strain value obtained at the current moment is used. To determine failure based on the values in the set of strain inflection points of a temporary body, and to define the set of failure criteria for strain inflection points of a temporary body:
[0068] in, The length of the set after dynamically inserting the strain inflection point of the temporary body. S126. Set the failure tolerance threshold. Merge temporary strain inflection point set With failure discrimination set The final set of judgments for volumetric strain inflection points is obtained as follows: S127. During the loading process of rock samples, local volumetric strain inflection points sometimes occur before the volumetric strain reaches its maximum value due to data fluctuations. Since subsequent damage calculations depend on the volumetric strain inflection point value, the existence of local volumetric strain inflection points will affect the damage calculation. To reduce the impact of local volumetric strain inflection points on the damage results, this paper incorporates the duration of the inflection point for screening. Assuming the duration of a suspected volumetric strain inflection point is... Set the inflection point duration to Add new judgment conditions:
[0069] Thus, the duration was not reached. The inflection point will be eliminated; S128. Select the values of the volumetric strain inflection points from the set:
[0070] S129. After calculating the inflection point value, the volumetric strain value is obtained in real time. Inflection point of volumetric strain Calculate the real-time damage of the rock sample: .
[0071] In specific implementation, as a preferred embodiment of the present invention, such as Figure 3 As shown, step S13 specifically includes: S131. The impairment of acoustic emission data is defined by the tangent angle of the dimensionless curve. The tangent angle between them is quantized and mapped to 0. The values between 1 and 2 represent the degree of damage; considering that the larger the angle, the more severe the increase in acoustic emission ringing number, an exponential quantization method is used (angle is converted to radians for calculation):
[0072] S132. To reduce the impact of local damage before rock sample failure on the overall results, this paper introduces the concept of a time factor to process the data. After the rock sample reaches the expansion point, the further away from the expansion point the damage occurs, the greater the probability of final failure. Therefore, a time factor is defined as follows:
[0073] In the formula, Indicates the current moment. This indicates the time when the inflection point of volumetric strain is reached. Indicates the predicted time of destruction; S133, Redefining Acoustic Emission Damage:
[0074] S134. For the dynamically acquired ring count sequence, calculate the cumulative ring count corresponding to the current moment in real time. , constructing a time-based For indexing purposes, Dynamic time window for index length Simultaneously, the cumulative ring count is stored in the cumulative ring count array. In the middle, the specific formula is as follows:
[0075]
[0076] S135, Regarding the current dynamic time window The velocity of the current time window is calculated using a difference method. and timing speed Store to velocity sequence In the middle, the specific formula is as follows:
[0077]
[0078] S136. After calculating the inflection point of the stable body strain from the strain data, the velocity series... Find the average velocity during the "uniform deformation" stage before the inflection point, and calculate the cumulative ringing sequence. Dimensionless transformation is performed, as shown in the following formula:
[0079]
[0080] S137. Regarding the obtained dimensionless time series... and the dimensionless value calculated at the current moment. , constructing a time-based For indexing purposes, Dynamic time window for index length :
[0081] S138, Regarding the current dynamic time window The slope of the time window at the current moment is calculated using a difference method. And calculate the slope arctangent value Ultimately Stored in angle sequence In the middle, the specific formula is as follows:
[0082]
[0083]
[0084] S139. Regarding the calculated angle sequence This is quantified as damage, as shown in the following formula: .
[0085] In specific implementation, as a preferred embodiment of the present invention, such as Figure 4 As shown, step S14 specifically includes: S141. After obtaining the volumetric strain matrix, the damage on a surface is considered as the accumulation of damage at all points, and the damage at a point is defined as follows:
[0086] In the formula, This represents the volumetric strain value at the expansion point. This represents the volumetric strain value at the current moment. The damage is calculated based on this formula. When the Poisson's ratio exceeds 0.5, it will be greater than 1. In this case, the rock sample is considered to be completely damaged, and its value is set to 1.
[0087] S142. In actual data processing, the volumetric strain data calculated from speckle data typically exhibits multiple "inflection points." This is considered to be caused by multiple localized damages, with the final damage result being the accumulation of these previous damages. As the loading pressure increases, the damage to the rock gradually intensifies. Damage closer to fracture is more likely to lead to rock instability and fracture. Therefore, when accumulating damage at multiple "inflection points," a time factor is used to control the weight of different damage types, characterizing the influence of time on instability and fracture. The time factor is defined as follows:
[0088] In the formula, Indicates the current moment. This indicates the time when the inflection point of volumetric strain is reached. Indicates the predicted time of destruction.
[0089] S143. Correct for damage, assuming that at a point, after the inflection point of volumetric strain, a defect occurs. Secondary local damage, then the damage at that point for:
[0090] S144. For dynamically acquired volumetric strain sequences and the volumetric strain value acquired at the current moment. , constructing a time-based As the index endpoint, Dynamic time window for index length The specific formula is as follows:
[0091] S145, For dynamic time windows Determine the local maximum value of the set according to the following criteria. Is it an inflection point (i.e., is the curve convex)? If the value is greater than 0, then the local maximum value point is the inflection point of the temporary body strain.
[0092] S146, Temporary body strain inflection points are dynamically stored in a set. In the middle, if the current set If not empty, then the volumetric strain value obtained at the current moment is used. To determine failure based on the values in the set of strain inflection points of a temporary body, and to define the set of failure criteria for strain inflection points of a temporary body:
[0093] in, The length of the set after dynamically inserting the strain inflection point of the temporary body; S147. Set the failure tolerance threshold. Realize the set of temporary body strain inflection points With failure discrimination set The merging of these results yields the final set of criteria for determining volumetric strain inflection points:
[0094] S148. Select the values of the volumetric strain inflection points from the set. And through the real-time acquired volumetric strain values Inflection point of volumetric strain Calculate the real-time damage of the rock sample:
[0095] For the The nth point, assuming the nth... The point appeared after the inflection point of the volumetric strain. Secondary local damage, then the damage at that point for:
[0096] S149, for dimensions of The volumetric strain matrix, the first Damage at each point is The current real-time damage of a speckle photograph is: .
[0097] In a specific implementation, as a preferred embodiment of the present invention, step S2 includes: S21. Define Dempster's composition rule for the DS evidence theory; S22. Modify the Dempster composition rule in response to the Zadeh paradox.
[0098] In a specific implementation, as a preferred embodiment of the present invention, step S21 specifically includes: S211, Regarding , The two mass functions m 1, m The synthesis rules for Dempster 2 are as follows:
[0099] in, K The normalization constant is ; S212, Regarding , A finite number of mass functions on m 1, m 2,…, m n The synthesis rules for Dempster are as follows:
[0100] in, K The normalization constant is .
[0101] In a specific implementation, as a preferred embodiment of the present invention, step S22 specifically includes: S221. Regarding the Zadeh paradox, the Dempster composition rule is rewritten as follows:
[0102] S222. Based on the rewritten Dempster composition rules in step S221, we have:
[0103] in, Let be the allocation function for the probability of evidence conflict. .
[0104] In specific implementation, as a preferred embodiment of the present invention, such as Figure 5 As shown, in step S3, the prediction of the destruction time is achieved through the inverse velocity method, specifically including: S31, From speckle pattern to orange alert From the start until the acoustic emission reaches the orange alert level. Finally, this interval is defined as the speckle-acoustic emission early warning interval, and the DS fusion damage results of the speckle-acoustic emission early warning interval constitute the damage time series. :
[0105] In the formula, Indicates the interval length. Indicates from arrive DS fusion damage results over the time period; S32. Based on the damage time series Obtain the inverse velocity sequence :
[0106] S33, Using the least squares method A single fitting is performed on the sequence data to obtain the fitted line, and the line is defined as... x The intersection of the axes is The predicted time of destruction is: .
[0107] In a preferred embodiment of this invention, the warning result in step S3 is characterized by a four-level warning system: blue, yellow, orange, and red. Based on the results of multi-source data fusion, the risk level of the rock sample can be classified into four levels: "blue," "yellow," "orange," and "red." The basis for this classification is as follows:
[0108] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.
Claims
1. A method for predicting and warning of uniaxial compressive rock instability based on multi-source data fusion, characterized in that, include: S1. Acquire monitoring data of rock samples during uniaxial compression loading and perform damage quantification on the data in real time, including: S11. The rock sample was monitored during uniaxial compression loading using DHDAS, PCI-2, and EOS 90D. Strain and acoustic emission data were acquired in real time using LabVIEW secondary development, and speckle data were acquired in real time using EOS Unity. S12. For the strain data acquired in real time, define strain damage quantification based on the volume strain inflection point; S13. For real-time acquired acoustic emission data, define the acoustic emission data quantification impairment based on the dimensionless curve of acoustic emission parameters, including: S131. Defining the impairment of acoustic emission data by the tangent angle of the dimensionless curve, quantizing the tangent angle between 45° and 90°, and mapping it to 0 to 1 to represent the degree of impairment; considering that the larger the angle, the more severe the increase in acoustic emission ringing, exponential quantization is adopted: S132. Define the time factor: In the formula, Indicates the current moment. This indicates the time when the inflection point of volumetric strain is reached. Indicates the predicted time of destruction; S133, Redefining Acoustic Emission Damage: S134. For the dynamically acquired ring count sequence, calculate the cumulative ring count corresponding to the current moment in real time. , build based on time As the index endpoint, Dynamic time window for index length Simultaneously, the cumulative ring count is stored in the cumulative ring count array. In the middle, the specific formula is as follows: S135, Regarding the current dynamic time window The velocity of the current time window is calculated using a difference method. and timing speed Store to velocity sequence In the middle, the specific formula is as follows: S136. After calculating the inflection point of the stable body strain from the strain data, the velocity series... Find the average velocity during the "uniform deformation" stage before the inflection point, and calculate the cumulative ringing sequence. Dimensionless transformation is performed, as shown in the following formula: S137. Regarding the obtained dimensionless time series... and the dimensionless value calculated at the current moment. , build based on time As the index endpoint, Dynamic time window for index length : S138, Regarding the current dynamic time window The slope of the time window at the current moment is calculated using a difference method. And calculate the slope arctangent value Ultimately Stored in an angle sequence In the middle, the specific formula is as follows: S139. Regarding the calculated angle sequence This is quantified as damage, as shown in the following formula: S14. For the speckle data acquired in real time, the damage is quantified based on the speckle data volume strain matrix defined by the optical flow method. S2. Based on the damage quantification results of multi-source monitoring data, the data are fused using the DS evidence theory; S3. Based on the single-index damage quantification results, a prediction of the damage time is given, and an early warning result is given according to the risk classification level based on the data fusion results.
2. The method for predicting and warning of uniaxial compressive rock instability based on multi-source data fusion according to claim 1, characterized in that, Step S12 specifically includes: S121, through axial strain and transverse strain The strain of the rock sample was calculated as follows: It is assumed that damage and accumulation in the rock sample begin after the volumetric strain expansion point. Considering the change in Poisson's ratio during loading, it is assumed that the rock sample is completely destroyed when the volumetric strain curve returns to zero, at which point the corresponding Poisson's ratio is 0.
5. Therefore, it is considered that the rock damage reaches its maximum value of 1 when the Poisson's ratio reaches 0.
5. S122. Based on the rock expansion point and Poisson's ratio, define the quantitative relationship between rock sample damage and volumetric strain: In the formula, This represents the volumetric strain value at the expansion point. This represents the volumetric strain value at the current moment. The damage is calculated based on this formula. When the Poisson's ratio exceeds 0.5, it will be greater than 1. In this case, the rock sample is considered to be completely damaged, and this value is set to 1. S123. For dynamically acquired volumetric strain sequences and the volumetric strain value acquired at the current moment. , build based on time As the index endpoint, Dynamic time window for index length The specific formula is as follows: S124, For dynamic time windows Determine the local maximum value of the set according to the following criteria. Is it an inflection point? If the value is greater than 0, then the local maximum value point is considered to be the inflection point of the temporary body strain. S125, Temporary body strain inflection points are dynamically stored in a set. In the middle, if the current set If not empty, then the volumetric strain value obtained at the current moment is used. To determine failure based on the values in the set of strain inflection points of a temporary body, and to define the set of failure criteria for strain inflection points of a temporary body: in, The length of the set after dynamically inserting the strain inflection point of the temporary body; S126. Set the failure tolerance threshold. Merge temporary strain inflection point set With failure discrimination set The final set of judgments for volumetric strain inflection points is obtained as follows: S127. Assume the duration of a suspected volumetric strain inflection point is... Set the inflection point duration to Add new judgment conditions: Thus, the duration was not reached. The inflection point will be eliminated; S128. Select the values of the volumetric strain inflection points from the set: S129. After calculating the inflection point value, the volumetric strain value is obtained in real time. Inflection point of volumetric strain Calculate the real-time damage of the rock sample: 。 3. The method for predicting and warning of uniaxial compressive rock instability based on multi-source data fusion according to claim 1, characterized in that, Step S14 specifically includes: S141. After obtaining the volumetric strain matrix, the damage on a surface is considered as the accumulation of damage at all points, and the damage at a point is defined as follows: In the formula, This represents the volumetric strain value at the expansion point. This represents the volumetric strain value at the current moment. The damage is calculated based on this formula. When the Poisson's ratio exceeds 0.5, it will be greater than 1. In this case, the rock sample is considered to be completely damaged, and its value is set to 1. S142. Define the time factor as: In the formula, Indicates the current moment. This indicates the time when the inflection point of volumetric strain is reached. Indicates the predicted time of destruction; S143. Correct for damage, assuming that at a point, after the inflection point of volumetric strain, a defect occurs. Secondary local damage, then the damage at that point for: S144. For dynamically acquired volumetric strain sequences and the volumetric strain value acquired at the current moment. , build based on time As the index endpoint, Dynamic time window for index length The specific formula is as follows: S145, For dynamic time windows Determine the local maximum value of the set according to the following criteria. Is it an inflection point? If the value is greater than 0, then the local maximum value point is the inflection point of the temporary body strain. S146, Temporary body strain inflection points are dynamically stored in a set. In the middle, if the current set If not empty, then the volumetric strain value obtained at the current moment is used. To determine failure based on the values in the set of strain inflection points of a temporary body, and to define the set of failure criteria for strain inflection points of a temporary body: in, The length of the set after dynamically inserting the strain inflection point of the temporary body; S147. Set the failure tolerance threshold. Realize the set of temporary body strain inflection points With failure discrimination set The merging of these results yields the final set of criteria for determining volumetric strain inflection points: S148. Select the values of the volumetric strain inflection points from the set. And through the real-time acquired volumetric strain values Inflection point of volumetric strain Calculate the real-time damage of the rock sample: For the The nth point, assuming the nth... The point appeared after the inflection point of the volumetric strain. Secondary local damage, then the damage at that point for: S149, for dimensions of The volumetric strain matrix, the first Damage at each point is The current real-time damage of a speckle photograph is: 。 4. The method for predicting and warning of uniaxial compressive rock instability based on multi-source data fusion according to claim 1, characterized in that, Step S2 includes: S21. Define Dempster's composition rule for the DS evidence theory; S22. Modify the Dempster composition rule in response to the Zadeh paradox.
5. The method for predicting and warning of uniaxial compressive rock instability based on multi-source data fusion according to claim 4, characterized in that, Step S21 specifically includes: S211, Regarding , The two mass functions m 1, m The synthesis rules for Dempster 2 are as follows: in, K The normalization constant is ; S212, Regarding , A finite number of mass functions on m 1, m 2,…, m n The synthesis rules for Dempster are as follows: in, K The normalization constant is .
6. The method for predicting and warning of uniaxial compressive rock instability based on multi-source data fusion according to claim 4, characterized in that, Step S22 specifically includes: S221. Regarding the Zadeh paradox, the Dempster composition rule is rewritten as follows: S222. Based on the rewritten Dempster composition rules in step S221, we have: in, f(A) This is the allocation function for the probability of evidence conflict.
7. The method for predicting and warning of uniaxial compressive rock instability based on multi-source data fusion according to claim 1, characterized in that, In step S3, the destruction time is predicted using the inverse velocity method, specifically including: S31, From speckle pattern to orange alert From the start until the acoustic emission reaches the orange alert level. Finally, this interval is defined as the speckle-acoustic emission early warning interval, and the DS fusion damage results of the speckle-acoustic emission early warning interval constitute the damage time series. : In the formula, Indicates the interval length. Indicates from arrive DS fusion damage results over the time period; S32. Based on the damage time series Obtain the inverse velocity sequence : S33, Using the least squares method A single fitting is performed on the sequence data to obtain the fitted line, and the line is defined as... x The intersection of the axes is The predicted time of destruction is: 。 8. The method for predicting and warning of uniaxial compressive rock instability based on multi-source data fusion according to claim 1, characterized in that, The warning results in step S3 are represented by a four-level warning system: blue, yellow, orange, and red.