Tunnel surrounding rock risk monitoring system and carrier device based on multi-source data fusion

Abstract

The present application belongs to the technical field of tunnel construction, and particularly relates to a tunnel surrounding rock risk monitoring system and carrier equipment based on multi-source data fusion, aiming to realize high-quality monitoring of tunnel surrounding rock safety risks, comprising a monitoring center, a data acquisition module, an information processing module, a risk assessment module and an intelligent early warning module. The data acquisition module collects multi-dimensional construction data and captures time at monitoring device points. The information processing module performs primary fusion on the multi-dimensional construction data to obtain a sequence fluctuation curve graph, and obtains a predicted difference value after data extraction. The risk assessment module performs fusion analysis on the predicted difference value based on the sequence fluctuation curve graph to obtain an estimated weight difference value, and obtains a variable fluctuation marker value according to the estimated weight difference value. The intelligent early warning module divides the surrounding rock risks according to the variable fluctuation marker value to obtain a surrounding rock risk grade, and obtains a surrounding rock prevention strategy according to the interval warning of the surrounding rock risk grade.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of tunnel construction, and particularly relates to a tunnel surrounding rock risk monitoring system based on multi-source data fusion and a carrier device. BACKGROUND

[0002] With the rapid advancement of infrastructure construction, the role of tunnel engineering in the fields of transportation and underground space development is becoming increasingly critical. However, the tunnel construction environment has high complexity and high uncertainty, and factors such as surrounding rock stability and geological condition mutation can easily cause safety risks, posing a serious threat to the safety of construction personnel and the progress of the project.

[0003] Traditional tunnel safety monitoring methods rely on single means, such as manual inspection or local sensor monitoring, which have limited data coverage and fragmented information, making it difficult to fully reflect the construction dynamics. At the same time, these methods lack systematic integration in data processing, making it difficult to capture the correlation between multiple factors, resulting in delayed risk early warning and insufficient evaluation accuracy, which cannot meet the safety control needs of complex tunnel engineering. SUMMARY

[0004] The purpose of the present application is to provide a tunnel surrounding rock risk monitoring system based on multi-source data fusion and a carrier device, aiming to achieve high-quality monitoring of tunnel surrounding rock safety risks.

[0005] To achieve the above-mentioned purpose, the technical scheme adopted by the present application is as follows:

[0006] The present application provides a tunnel surrounding rock risk monitoring system based on multi-source data fusion, comprising a monitoring center, a data acquisition module, an information processing module, a risk assessment module and an intelligent early warning module. The monitoring center is connected with the data acquisition module, the information processing module, the risk assessment module and the intelligent early warning module. The data acquisition module is connected with the information processing module. The information processing module is connected with the risk assessment module. The risk assessment module is also connected with the intelligent early warning module. The data acquisition module is configured to collect multi-dimensional construction data and capture time at the monitoring device point. The information processing module is configured to obtain a sequence fluctuation curve graph by primary fusion of the multi-dimensional construction data, and obtain a prediction difference value by data extraction of the sequence fluctuation curve graph. The risk assessment module is configured to obtain an estimated weight difference value by fusion analysis of the prediction difference value based on the sequence fluctuation curve graph, and obtain a variable fluctuation marker value by fluctuation classification according to the estimated weight difference value. The intelligent early warning module is configured to obtain a surrounding rock risk grade by surrounding rock risk division according to the variable fluctuation marker value, and obtain a surrounding rock prevention strategy by interval early warning according to the surrounding rock risk grade.

[0007] The multi-dimensional construction data and the time of capturing are collected at the monitoring device points, which comprises: obtaining a tunnel construction monitoring scheme, determining the monitoring device points according to the tunnel construction monitoring scheme, collecting the multi-dimensional construction data based on the monitoring device points, and recording the corresponding data capturing time.

[0008] The multi-dimensional construction data is preliminarily fused to obtain a sequence fluctuation curve, which comprises: removing the abnormal data from the multi-dimensional construction data to obtain cleaned multi-dimensional data, combining the cleaned multi-dimensional data in time sequence based on the capturing time to obtain a multi-dimensional data sequence, and integrating the multi-dimensional data sequence into a two-dimensional rectangular coordinate system with the capturing time as the horizontal axis to form the sequence fluctuation curve.

[0009] The sequence fluctuation curve is subjected to data extraction to obtain a predicted difference value, which comprises: marking a sequence measurement point on the sequence fluctuation curve according to the capturing time, performing difference fusion on the sequence fluctuation curve based on the sequence measurement point to obtain the predicted difference value. The predicted difference value is obtained by removing the abnormal data from the multi-dimensional construction data to obtain cleaned multi-dimensional data, combining the cleaned multi-dimensional data in time sequence based on the capturing time to obtain a multi-dimensional data sequence, and integrating the multi-dimensional data sequence into a two-dimensional rectangular coordinate system with the capturing time as the horizontal axis to form the sequence fluctuation curve. The predicted difference value is obtained by removing the abnormal data from the multi-dimensional construction data to obtain cleaned multi-dimensional data, combining the cleaned multi-dimensional data in time sequence based on the capturing time to obtain a multi-dimensional data sequence, and integrating the multi-dimensional data sequence into a two-dimensional rectangular coordinate system with the capturing time as the horizontal axis to form the sequence fluctuation curve. The predicted difference value is obtained by removing the abnormal data from the multi-dimensional construction data to obtain cleaned multi-dimensional data, combining the cleaned multi-dimensional data in time sequence based on the capturing time to obtain a multi-dimensional data sequence, and integrating the multi-dimensional data sequence into a two-dimensional rectangular coordinate system with the capturing time as the horizontal axis to form the sequence fluctuation curve. The predicted difference value is obtained by removing the abnormal data from the multi-dimensional construction data to obtain cleaned multi-dimensional data, combining the cleaned multi-dimensional data in time sequence based on the capturing time to obtain a multi-dimensional data sequence, and integrating the multi-dimensional data sequence into a two-dimensional rectangular coordinate system with the capturing time as the horizontal axis to form the sequence fluctuation curve.

[0010] The sequence fluctuation curve is subjected to data extraction to obtain a predicted difference value, which comprises: marking a sequence measurement point on the sequence fluctuation curve according to the capturing time, performing difference fusion on the sequence fluctuation curve based on the sequence measurement point to obtain the predicted difference value. The sequence fluctuation curve is subjected to data extraction to obtain a predicted difference value, which comprises: marking a sequence measurement point on the sequence fluctuation curve according to the capturing time, performing difference fusion on the sequence fluctuation curve based on the sequence measurement point to obtain the predicted difference value. The sequence fluctuation curve is subjected to data extraction to obtain a predicted difference value, which comprises: marking a sequence measurement point on the sequence fluctuation curve according to the capturing time, performing difference fusion on the sequence fluctuation curve based on the sequence measurement point to obtain the predicted difference value. The sequence fluctuation curve is subjected to data extraction to obtain a predicted difference value, which comprises: marking a sequence measurement point on the sequence fluctuation curve according to the capturing time, performing difference fusion on the sequence fluctuation curve based on the sequence measurement point to obtain the predicted difference value. The sequence fluctuation curve is subjected to data extraction to obtain a predicted difference value, which comprises: marking a sequence measurement point on the sequence fluctuation curve according to the capturing time, performing difference fusion on the sequence fluctuation curve based on the sequence measurement point to obtain the predicted difference value. The sequence fluctuation curve is subjected to data extraction to obtain a predicted difference value, which comprises: marking a sequence measurement point on the sequence fluctuation curve according to the capturing time, performing difference fusion on the sequence fluctuation curve based on the sequence measurement point to obtain the predicted difference value. The sequence fluctuation curve is subjected to data extraction to obtain a predicted difference value, which comprises: marking a sequence measurement point on the sequence fluctuation curve according to the capturing time, performing difference fusion on the sequence fluctuation curve based on the sequence measurement point to obtain the predicted difference value. The sequence fluctuation curve is subjected to data extraction to obtain a predicted difference value, which comprises: marking a sequence measurement point on the sequence fluctuation curve according to the capturing time, performing difference fusion on the sequence fluctuation curve based on the sequence measurement point to obtain the predicted difference value.

[0011] The sequence fluctuation curve is subjected to data extraction to obtain a predicted difference value, which comprises: marking a sequence measurement point on the sequence fluctuation curve according to the capturing time, performing difference fusion on the sequence fluctuation curve based on the sequence measurement point to obtain the predicted difference value. The sequence fluctuation curve is subjected to data extraction to obtain a predicted difference value, which comprises: marking a sequence measurement point on the sequence fluctuation curve according to the capturing time, performing difference fusion on the sequence fluctuation curve based on the sequence measurement point to obtain the predicted difference value.

[0012] The sequence fluctuation curve is subjected to data extraction to obtain a predicted difference value, which comprises: marking a sequence measurement point on the sequence fluctuation curve according to the capturing time, performing difference fusion on the sequence fluctuation curve based on the sequence measurement point to obtain the predicted difference value.

[0013] The interval early warning according to the surrounding rock risk level obtains a surrounding rock prevention strategy, including: dividing warning intervals according to the surrounding rock risk level, determining corresponding risk early warning distances; reserving emergency intervals on the weight difference mark graph based on the risk early warning distances; and performing early warning judgment and generating a surrounding rock prevention strategy according to the reserved emergency intervals.

[0014] The multi-dimensional construction data include the face surrounding rock deformation data, the support deformation data and the comprehensive geological condition; and the monitoring device point is an information acquisition device placement point determined according to a tunnel construction monitoring scheme.

[0015] The application further provides a carrier device comprising a memory and a processor; the memory is connected with the processor, and the memory stores the tunnel surrounding rock risk monitoring system based on multi-source data fusion as described above.

[0016] Compared with the prior art, the application has the beneficial effects that:

[0017] 1. The tunnel surrounding rock risk monitoring system based on multi-source data fusion provided by the application integrates multi-dimensional data such as face surrounding rock deformation, support structure state and comprehensive geological condition through the data acquisition module, effectively filters noise data in combination with abnormal elimination processing, ensures the accuracy of the analysis basis, breaks through the limitation of traditional single monitoring means through multi-source data fusion, and realizes all-around coverage of the tunnel construction environment.

[0018] 2. The sequence fluctuation curve diagram is used to visually present the time series data change trend, the data fluctuation characteristics are captured through quantitative indexes such as predicted difference value and estimated weight difference value, and the fine classification of the surrounding rock risk is realized through span period dynamic division and critical threshold analysis, so as to avoid subjective judgment deviation and improve the reliability of the evaluation result.

[0019] 3. The surrounding rock risk level is divided based on the variable fluctuation mark value, different warning intervals and early warning distances are matched, the emergency response time is reserved for the high-risk area, the targeted prevention strategy is generated, the passive response is changed to active prevention and control, the accident occurrence probability is significantly reduced, and the safety of construction personnel is ensured. And through the modular design, the full-process automation of data acquisition, processing, evaluation and early warning is realized, the manual intervention cost is reduced; the real-time monitoring and hierarchical early warning mechanism helps the management personnel to quickly locate the risk point, reasonably allocates resources, balances the construction progress and safety control, and improves the overall engineering efficiency. BRIEF DESCRIPTION OF DRAWINGS

[0020] Figure 1 is a structural schematic diagram of a tunnel surrounding rock risk monitoring system based on multi-source data fusion provided by the application.

[0021] Wherein, 1 is a monitoring center, 2 is a data acquisition module, 3 is an information processing module, 4 is a risk assessment module, and 5 is an intelligent early warning module. DETAILED DESCRIPTION

[0022] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work fall within the scope of the present application.

[0023] The embodiments of the present application provide a tunnel surrounding rock risk monitoring system based on multi-source data fusion, which is exemplarily shown in Figure 1 The tunnel surrounding rock risk monitoring system comprises a monitoring center 1, a data acquisition module 2, an information processing module 3, a risk assessment module 4, and an intelligent early warning module 5. The monitoring center 1 is connected with the data acquisition module 2, the information processing module 3, the risk assessment module 4, and the intelligent early warning module 5 respectively. The data acquisition module 2 is connected with the information processing module 3. The information processing module 3 is connected with the risk assessment module 4. The risk assessment module 4 is further connected with the intelligent early warning module 5.

[0024] The data acquisition module 2 is configured to collect multi-dimensional construction data and capture time at the monitoring device point. The information processing module 3 is configured to obtain a sequence fluctuation curve graph by primary fusion of the multi-dimensional construction data, and obtain a predicted difference value by data extraction of the sequence fluctuation curve graph. The risk assessment module 4 is configured to obtain an estimated weight difference value by fusion analysis of the predicted difference value based on the sequence fluctuation curve graph, and obtain a variable fluctuation marker value by fluctuation classification according to the estimated weight difference value. The intelligent early warning module 5 is configured to obtain a surrounding rock risk grade by surrounding rock risk division according to the variable fluctuation marker value, and obtain a surrounding rock prevention strategy by interval early warning according to the surrounding rock risk grade.

[0025] Collecting multi-dimensional construction data and capturing time at the monitoring device point comprises: obtaining a tunnel construction monitoring scheme, determining the monitoring device point according to the tunnel construction monitoring scheme, collecting multi-dimensional construction data based on the monitoring device point, and recording the corresponding data capture time.

[0026] Exemplarily, the tunnel construction monitoring scheme represents a specific construction plan determined before tunnel excavation, including preliminary investigation and exploration, scheme design, technical design, construction drawing design, construction standard details, monitoring frequency, monitoring content, and construction preparation.

[0027] The monitoring device point is a placement point of an information collection device determined according to the obtained tunnel construction monitoring scheme. A corresponding information collection device is installed at the monitoring device point, such as an optical fiber sensor, a monitoring device, or a strain gauge. In the embodiment of the present application, the optical fiber sensor and the strain gauge are used in combination. The optical fiber sensor is used for distributed monitoring, and the strain gauge is used for accurate measurement of key positions, to jointly collect detailed monitoring device point information. The key positions refer to weak points of construction or positions with greater construction difficulty in the tunnel construction monitoring scheme.

[0028] For example, the multi-dimensional construction data includes face surrounding rock deformation data, support deformation data, and comprehensive geological conditions. The face surrounding rock deformation data includes surrounding rock displacement data, surrounding rock internal force data, and surrounding rock stress data. The support deformation data includes support structure deformation data, such as deformation of an anchor rod, shotcrete, and a steel arch support structure, and is collected by a strain gauge or an optical fiber sensor installed on the support structure. The comprehensive geological conditions include geological structure information, stratum lithology information, fault information, joint number information, fracture development information, underground water level data, environmental temperature and humidity data, and construction activity data. The obtained multi-dimensional construction data is associated with the corresponding monitoring device point.

[0029] The multi-dimensional construction data is preliminarily fused to obtain a sequence fluctuation curve diagram, including:

[0030] The multi-dimensional construction data is subjected to abnormality elimination to obtain cleaned multi-dimensional data. Abnormality elimination refers to abnormal value screening of the collected multi-dimensional construction data. Abnormal values deviating from a normal range are eliminated. For example, a specification threshold is set. Multi-dimensional construction data meeting the specification threshold is retained, and multi-dimensional construction data not meeting the specification threshold is deleted, to obtain cleaned multi-dimensional data. For example, the specification threshold of surrounding rock displacement data is a1 to a2. If the collected surrounding rock displacement data is greater than a2, it is an abnormal value, and the collected surrounding rock displacement data is deleted.

[0031] The cleaned multi-dimensional data is subjected to time sequence combination based on a capture time to obtain a multi-dimensional data sequence.

[0032] The obtained cleaned multi-dimensional data is subjected to time sequence combination based on a capture time to obtain a multi-dimensional data sequence, and time statistics is performed on the multi-dimensional data sequence to obtain a combination time period, indicating the length of the capture time contained in the multi-dimensional data sequence. The time sequence combination means that the cleaned multi-dimensional data of the same type is combined into a data sequence with a time sequence according to the time sequence of the capture time.

[0033] For example, because the multi-dimensional construction data includes face surrounding rock deformation data, support deformation data, and comprehensive geological conditions, as a possible implementation manner, the multi-dimensional data sequence includes face surrounding rock deformation sequence, support deformation sequence, and comprehensive geological sequence.

[0034] The multi-dimensional data sequence is integrated into a two-dimensional rectangular coordinate system with the capture time as the horizontal axis to form a sequence fluctuation curve diagram.

[0035] A two-dimensional rectangular coordinate system with respect to the capture time is constructed, the obtained multi-dimensional data sequence is uploaded to the two-dimensional rectangular coordinate system, and curve integration is performed on the obtained multi-dimensional data sequence to obtain a multi-dimensional sequence curve. The curve integration means that, according to the order of the capture time, the multi-dimensional cleaning data in the multi-dimensional data sequence is marked at the corresponding position of the two-dimensional rectangular coordinate system according to the position information corresponding to the capture time, and a smooth curve is used to connect two adjacent multi-dimensional cleaning data to obtain the multi-dimensional sequence curve. The two-dimensional rectangular coordinate system including the multi-dimensional sequence curve is denoted as a sequence fluctuation curve diagram. In particular, according to the sequence of the surrounding rock of the working face, the sequence of the support deformation, and the comprehensive geological sequence included in the multi-dimensional data sequence, the multi-dimensional sequence curve includes the surrounding rock deformation curve of the working face, the support deformation curve, and the comprehensive geological curve, which are all represented in the same sequence fluctuation curve diagram.

[0036] According to the obtained capture time, coordinate marking is performed on the sequence fluctuation curve diagram to obtain a sequence measurement point. The coordinate marking means that, in the sequence fluctuation curve diagram, the intersection point of the horizontal axis and the multi-dimensional sequence curve is denoted as the sequence measurement point, that is, the intersection point of the capture time and the multi-dimensional sequence curve is denoted as the sequence measurement point. The obtained sequence measurement point is denoted as t, which represents the number of the capture time on the horizontal axis, t = 1, 2, 3, …, v1, and v1 is a positive integer.

[0037] The data extraction is performed on the sequence fluctuation curve diagram to obtain a prediction difference value, including: according to the capture time, the sequence measurement point is marked on the sequence fluctuation curve diagram; based on the sequence measurement point, the sequence fluctuation curve diagram is divided into a plurality of sub-sequence fluctuation curve diagrams according to the capture time, and the sub-sequence fluctuation curve diagrams are sequentially arranged according to the capture time; based on the sequence measurement point, the difference integration is performed on the sequence fluctuation curve diagram according to the capture time to obtain the prediction difference value. The prediction difference value is the maximum value of the sequence measurement point on the sequence fluctuation curve diagram, the minimum value of the sequence measurement point on the sequence fluctuation curve diagram, the value at the sequence measurement point, and m represents the number of the sequence fluctuation curve diagram, m = 1, 2, 3, …, v2, and v2 is a positive integer.

[0038] The fusion analysis is performed on the prediction difference value based on the sequence fluctuation curve diagram to obtain an estimated weight difference value, including: according to the prediction difference value, the weight difference estimation is performed according to the capture time to obtain a fusion weight value; the span period is set on the sequence fluctuation curve diagram, and based on the span period and the fusion weight value, the intersection point of the sequence measurement point is estimated to obtain the estimated weight difference value. ​​​​denotes a value at the t-th sequence measurement point in the k-th span period, denotes corresponding fusion weight value, k denotes the number of span period in the sequence fluctuation curve, k = 1, 2, 3, … v3, v3 is a positive integer.

[0039] The span period is a fixed length time period selected according to the capture time, including a plurality of capture times, and the time period can be flexibly changed, but the multi-dimensional construction data corresponding to the sequence fluctuation curve belongs to the same time period, and only after this processing and analysis is completed, the length of the time period can be changed next time. The obtained estimated weight difference value is associated with the corresponding span period, and the estimated weight difference value is marked in the sequence fluctuation curve at the corresponding span period.

[0040] As a possible implementation manner, if it is desired to observe the fluctuation change of the multi-dimensional sequence curve in more detail, the span period can be set to be short enough, such as only including one or two capture periods.

[0041] As a possible implementation manner, the multi-dimensional sequence curve includes a surrounding rock deformation curve of a working face, a support deformation curve, and a comprehensive geological curve, and the estimated weight difference value is obtained according to the corresponding curve. For example, for the surrounding rock deformation curve of the working face, a plurality of estimated weight difference values can be obtained, such as a plurality of estimated weight difference values corresponding to the surrounding rock displacement data, which are estimated weight difference values of span periods corresponding to all capture times of the multi-dimensional data sequence. The fluctuation size of the multi-dimensional construction data can be obtained by comparing the estimated weight difference values of different span periods in the sequence fluctuation curve, such as obtaining the surrounding rock displacement fluctuation in the combined time period by observing the estimated weight difference values of the surrounding rock displacement curve corresponding to the surrounding rock displacement data. According to the fluctuation of the surrounding rock displacement, the fluctuation level can be divided, and the risk level of the monitoring device point can be obtained according to the divided fluctuation level and each multi-dimensional construction data, which is used for intelligent early warning.

[0042] According to the estimated weight difference value, the variable fluctuation marking value is obtained by converting the estimated weight difference value into a weight difference identification map based on the span period.

[0043] For example, according to the number order of the span period, a two-dimensional rectangular coordinate system is constructed, the horizontal axis of the constructed two-dimensional rectangular coordinate system is the span period, and the vertical axis represents the estimated weight difference value. The estimated weight difference value corresponding to each multi-dimensional sequence curve in the sequence fluctuation curve is marked at the corresponding span period, for example, a plurality of estimated weight difference values corresponding to the surrounding rock displacement data are marked at the corresponding position of the span period in the two-dimensional rectangular coordinate system, to obtain a weight difference identification map. It should be understood that for each multi-dimensional sequence curve, a weight difference identification map can be generated;

[0044] Set a critical threshold axis on the weight difference labeling chart, divide the weight difference fluctuation segment, and assign values ​​to obtain the variable fluctuation label value.

[0045] A critical threshold axis is set for the obtained weight difference labeling chart. This axis includes a critical upper limit and a critical lower limit, which are straight lines parallel to the horizontal axis of the weight difference labeling chart and can be shifted vertically within the chart. The distance between the critical upper and lower limits is a pre-defined threshold range used to distinguish different fluctuation ranges in the estimated weight difference values. Corresponding risk levels should be determined for different fluctuation ranges. The obtained critical threshold axis is uploaded to the weight difference labeling chart, and the chart is dynamically divided using this axis to obtain weight difference fluctuation segments. Values ​​are then assigned to these segments to obtain variable fluctuation marker values.

[0046] It needs further explanation that, in the specific implementation process, dynamic segmentation refers to differentiating the degree of fluctuation by moving the critical threshold axis to distinguish the estimated weight difference values ​​falling on different critical threshold axes. This results in different levels of weight difference fluctuation segments, which are then assigned different values. This facilitates risk level classification and allows for more detailed observation of surrounding rock deformation, support deformation, and environmental changes at the monitoring device points. This enables more accurate intelligent risk warnings. The specific process includes:

[0047] The obtained critical threshold axis is classified into levels based on the weight difference identification map to obtain axis segment classification. The level classification represents the distance between the critical upper limit and critical lower limit of the critical threshold axis determined according to the weight difference identification map, and the number of levels that can be divided into the weight difference identification map, which is the axis segment classification. For example, if the axis segment classification can be obtained as 6 according to the weight difference identification map, then the critical threshold axis can be moved up and down in the weight difference identification map to divide the estimated weight difference value in the weight difference identification map into 6 levels.

[0048] According to the obtained axial segment classification, the critical threshold axis is moved by classification, and the weight fluctuation segment is obtained, wherein the classification movement means that according to the axial segment classification, the estimated weight value is classified in the weight identification map according to the number of axial segment classification, and is recorded as the weight fluctuation segment; in particular, the weight fluctuation segment falling in each axial segment classification is recorded as the fluctuation level of the same level, for example, as mentioned above, the axial segment classification is 6, and the corresponding weight fluctuation segment has 6 segments, the estimated weight value falling in the first segment is recorded as the first weight fluctuation segment, the estimated weight value falling in the second segment is recorded as the second weight fluctuation segment, and so on, and the estimated weight value falling in the sixth segment is recorded as the sixth weight fluctuation segment; further, for the weight identification map composed of the estimated weight value corresponding to each multi-dimensional construction data, there is a corresponding critical threshold axis, but for the same tunnel construction safety risk assessment, the critical threshold axis of the same type of data is the same, which is convenient for standardization and standardization comparison.

[0049] According to the weight fluctuation segment divided in the weight identification map, the weight fluctuation segment of different levels is valued, and the variable fluctuation mark value is obtained; that is, each level of the weight fluctuation segment is given a numerical value, and the estimated weight value is valued from small to large, that is, the weight fluctuation segment corresponding to the smallest estimated weight value is valued the smallest, and the subsequent values are increased in turn, for example, according to the above known 6 weight fluctuation segments, the first weight fluctuation segment has the smallest estimated weight value, then the first weight fluctuation segment is valued 1, the second weight fluctuation segment is valued 2, the third weight fluctuation segment is valued 3, and so on; exemplarily, as long as the valuation of the same tunnel construction safety risk assessment is the same, only the valuation of the next tunnel construction safety risk assessment can be changed.

[0050] For each kind of multi-dimensional construction data, the corresponding variable fluctuation mark value can be obtained, if all types of variable fluctuation mark values are integrated and analyzed within a span period, the risk fluctuation value at the span period can be obtained, and the risk level is divided according to the risk fluctuation value, that is, the surrounding rock risk level of the monitoring device point within a period of time.

[0051] According to the variable fluctuation mark value, the surrounding rock risk classification is divided to obtain the surrounding rock risk level, which comprises: based on the span period, the variable fluctuation mark value is divided into items, and the summary fluctuation coefficient is obtained.

[0052] Exemplarily, the obtained summary fluctuation coefficient is marked as HZ, for example, at the kth span period, the variable fluctuation marked values of the surrounding rock deformation data of the working face, the support deformation data and the comprehensive geological conditions are z1, z2, z3, z4, z5, z6 and z7, and the summary fluctuation coefficient of the kth span period is HZ=z1+z2+z3+z4+z5+z6+z7. Further, the summary fluctuation coefficient is a value of quantified fluctuation, which is a value of assignment of fluctuation ranges of all multi-dimensional construction data in the span period, and is a coefficient value for risk grade division.

[0053] A separation threshold is set, and the surrounding rock risk grade is obtained by grading the summary fluctuation coefficient through the separation threshold.

[0054] Exemplarily, the separation threshold is set according to the obtained summary fluctuation coefficient, the separation threshold includes a first separation value and a second separation value, and the first separation value is smaller than the second separation value. The first separation value is marked as F1, and the second separation value is marked as F2. The threshold for grading the summary fluctuation coefficient is pre-set, which is set according to the construction standard details of the tunnel construction monitoring scheme and conforms to the threshold of the industry construction standard.

[0055] The surrounding rock risk grade is obtained by grading the summary fluctuation coefficient through the separation threshold, and the surrounding rock risk grade includes a first risk surrounding rock, a second risk surrounding rock and a third risk surrounding rock.

[0056] It needs to be further explained that, in the specific implementation process, the surrounding rock grading means that the summary fluctuation coefficient is divided into corresponding range risk grades through the separation threshold, that is, the surrounding rock risk grade, which is used for intelligent early warning of different risk grades to improve the safety of tunnel construction. The specific process includes:

[0057] The summary fluctuation coefficient is compared through the separation threshold. When HZ>F2, the obtained surrounding rock risk grade corresponding to the summary fluctuation coefficient is recorded as the first risk surrounding rock. The corresponding estimated weight difference value is relatively large, and the risk grade of the first risk surrounding rock is the highest.

[0058] When F1≤HZ≤F2, the surrounding rock risk grade corresponding to the summary fluctuation coefficient is recorded as the second risk surrounding rock. The summary fluctuation coefficient of this part is smaller than that of the first risk surrounding rock, that is, the risk grade of the second risk surrounding rock is weaker than that of the first risk surrounding rock.

[0059] When HZ<F1, the surrounding rock risk grade corresponding to the summary fluctuation coefficient is recorded as the third risk surrounding rock. The summary fluctuation coefficient of this part is smaller than that of the second risk surrounding rock, that is, the risk grade of the third risk surrounding rock is weaker than that of the second risk surrounding rock.

[0060] The surrounding rock prevention strategy obtained according to the interval early warning according to the surrounding rock risk level comprises: dividing the warning interval according to the surrounding rock risk level, determining the corresponding risk early warning distance; reserving an emergency interval on the weight difference identification map based on the risk early warning distance; determining the early warning according to the reserved emergency interval and generating the surrounding rock prevention strategy.

[0061] The risk early warning distance comprises long-distance early warning, medium-distance early warning and short-distance early warning. The warning interval according to the surrounding rock risk level comprises a long-distance warning interval, a medium-distance warning interval and a short-distance warning interval. In this embodiment, the long-distance warning interval is a range of more than 200 meters from the monitoring device point, the medium-distance warning interval is a range of 30 to 200 meters from the monitoring device point, and the short-distance warning interval is a range of within 30 meters from the monitoring device point.

[0062] According to the obtained warning interval matching the corresponding surrounding rock risk level, the risk early warning distance corresponding to the first risk surrounding rock is recorded as long-distance early warning, indicating that when the long-distance warning interval from the first risk surrounding rock is monitored, the first risk warning is performed. The risk early warning distance corresponding to the second risk surrounding rock is recorded as medium-distance early warning, indicating that when the medium-distance warning interval from the second risk surrounding rock is monitored, the second risk warning is performed. The risk early warning distance corresponding to the third risk surrounding rock is recorded as short-distance early warning, indicating that when the medium-distance warning interval from the third risk surrounding rock is monitored, the third risk warning is performed. That is, for the monitoring device point with higher risk level, early warning should be performed at a longer distance to leave enough time for risk prevention, so as to improve the safety of tunnel construction and reduce the possibility of risk occurrence.

[0063] For the surrounding rock prevention strategy, the first risk warning is to broadcast the risk forecast to the entire tunnel construction range when the long-distance early warning range is reached, play the corresponding emergency evacuation plan, and carry out early warning rescue work. The second risk warning is to broadcast the risk early warning in the radiation range of the medium-distance range early warning, play the evacuation plan, and perform warning work on other non-early warning areas of the medium-distance range early warning. While normal construction is carried out, the evacuation preparation work is also done.

[0064] The risk early warning distance is reserved on the weight difference identification map, that is, the weight difference identification map at the monitoring device point is obtained, and the weight difference identification map is time-reserved according to the obtained risk early warning distance to obtain the reserved emergency interval.

[0065] Further, the time reservation represents that the risk warning distance is used to divide the warning time of the spread difference map, that is, the monitoring device points of different risk warning distances are reserved for safety prevention time, including generating long reservation time, medium reservation time and short reservation time according to the risk warning distance. Generating long reservation time according to risk warning distance means that at the monitoring device point of long distance warning span period, the time period of long reservation time is selected forward. Generating medium reservation time according to risk warning distance is at the monitoring device point of medium distance warning span period, the time period of medium reservation time is selected forward. Generating short reservation time according to risk warning distance is at the monitoring device point of short distance warning span period, the time period of short reservation time is selected forward. Such hierarchical classification can leave a long enough and far enough place for the monitoring device point with higher risk level to prepare for preventive measures, so as to quickly solve the high risk warning and find the monitoring point with high risk level of surrounding rock deformation in the shortest time to prevent safety.

[0066] According to the obtained reserved emergency interval, the monitoring device point is judged, and the surrounding rock prevention strategy is generated according to the warning judgment process. In the specific implementation process, the warning judgment means that the safety inspection is carried out according to the determined reserved emergency interval, risk warning distance and surrounding rock risk level, and the emergency measures in the inspection process are recorded to generate the corresponding surrounding rock prevention strategy, and according to the recorded surrounding rock prevention strategy, the risk prevention of the surrounding rock risk level of the next span period is carried out, which can prepare in advance to deal with the possible surrounding rock deformation risk and reduce the safety risk of tunnel construction.

[0067] The application also provides a carrier device comprising a memory and a processor. The memory is connected with the processor, and the memory stores the tunnel surrounding rock risk monitoring system based on multi-source data fusion as above. Since the memory stores the tunnel surrounding rock risk monitoring system based on multi-source data fusion, and the tunnel surrounding rock risk monitoring system provided by the application integrates multi-dimensional data such as surrounding rock deformation, support structure state and comprehensive geological conditions of the working face through the data acquisition module, and effectively filters noise data through abnormality elimination processing, the accuracy of the analysis basis can be ensured. Multi-source data fusion breaks through the limitations of traditional single monitoring method and realizes all-round coverage of the tunnel construction environment. The sequence fluctuation curve diagram is used to visually present the time series data change trend, the data fluctuation characteristics are captured through quantitative indexes such as predicted difference value and estimated spread value, and the span period dynamic division and critical threshold analysis are combined, which helps to realize fine classification of surrounding rock risk, avoid subjective judgment deviation, and improve the reliability of the evaluation result. In addition, the surrounding rock risk level is divided based on the variable fluctuation marker value, different warning intervals and warning distances are matched, the emergency response time is reserved for high-risk areas, the targeted prevention strategy is generated, the passive response is changed to active prevention and control, the accident probability is significantly reduced, and the safety of construction personnel is ensured.

[0068] In the description of the present specification, specific features, structures, materials or characteristics can be combined in any one or more embodiments or examples in an appropriate manner.

[0069] The above merely describes specific embodiments of the present application, but the protection scope of the present application is not limited thereto, and any person skilled in the art can easily think of changes or replacements within the technical range disclosed by the present application, which should be covered within the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. A tunnel surrounding rock risk monitoring system based on multi-source data fusion, characterized in that, The application relates to a tunnel surrounding rock risk monitoring system based on multi-source data fusion. The system comprises a monitoring center (1), a data acquisition module (2), an information processing module (3), a risk assessment module (4), and an intelligent early warning module (5). The monitoring center (1) is connected to the data acquisition module (2), the information processing module (3), the risk assessment module (4), and the intelligent early warning module (5), respectively. The data acquisition module (2) is connected to the information processing module (3), the information processing module (3) is connected to the risk assessment module (4), and the risk assessment module (4) is also connected to the intelligent early warning module (5). The data acquisition module (2) is configured to: acquire multidimensional construction data and capture time at the monitoring device point. The information processing module (3) is configured to: process multidimensional construction data. The process involves primary fusion to obtain a sequence fluctuation curve, followed by data extraction to obtain prediction difference values. This primary fusion of multidimensional construction data to obtain the sequence fluctuation curve includes: anomaly removal to clean the multidimensional construction data; time-series combination of the cleaned multidimensional data based on the capture time to obtain a multidimensional data sequence; and integration of the multidimensional data sequence into a two-dimensional Cartesian coordinate system with the capture time as the horizontal axis to form the sequence fluctuation curve. Data extraction to obtain prediction difference values ​​from the sequence fluctuation curve includes: marking sequence measurement points on the sequence fluctuation curve based on the capture time; and processing the sequence fluctuation curve based on the sequence measurement points according to: Differential fusion is performed to obtain the predicted difference value; where, To predict the difference value, This represents the maximum value of the measured points on the sequence fluctuation curve. This represents the minimum value of the measured points on the sequence fluctuation curve. The value at the measurement point of the sequence is represented by m, which represents the number of the sequence fluctuation curve, m = 1, 2, 3, ... v2, where v2 is a positive integer; the risk assessment module (4) is configured to: perform fusion analysis on the predicted difference value based on the sequence fluctuation curve to obtain the estimated weight difference value, and perform fluctuation classification according to the estimated weight difference value to obtain the variable fluctuation label value; wherein, the fusion analysis on the predicted difference value based on the sequence fluctuation curve to obtain the estimated weight difference value includes: according to the predicted difference value according to: The fusion weight value is obtained by estimating the weight difference; a span period is set on the sequence fluctuation curve, and the intersection point of the sequence measurement points is estimated based on the span period and the fusion weight value to obtain the estimated weight difference value. ;in, , This represents the value at the t-th sequence measurement point within the k-th span period. express The corresponding fusion weight value k represents the number of span periods in the sequence fluctuation curve, k = 1, 2, 3, …, v3, v3 is a positive integer; the variable fluctuation mark value obtained by performing fluctuation grading according to the estimated weight difference value comprises: converting the estimated weight difference value into a weight difference mark graph based on the span period; setting a critical threshold axis on the weight difference mark graph to divide the weight difference fluctuation section and obtain the variable fluctuation mark value; the intelligent early warning module (5) is configured to: perform surrounding rock risk division according to the variable fluctuation mark value to obtain a surrounding rock risk level, and perform interval early warning according to the surrounding rock risk level to obtain a surrounding rock prevention strategy; wherein the surrounding rock risk level obtained by performing surrounding rock risk division according to the variable fluctuation mark value comprises: performing item fusion on the variable fluctuation mark value based on the span period to obtain a summary fluctuation coefficient; setting a separation threshold to grade the summary fluctuation coefficient through the separation threshold to obtain the surrounding rock risk level; the surrounding rock prevention strategy obtained by performing interval early warning according to the surrounding rock risk level comprises: dividing a warning interval according to the surrounding rock risk level, determining a corresponding risk early warning distance; reserving an emergency interval on the weight difference mark graph based on the risk early warning distance; and performing early warning judgment according to the reserved emergency interval and generating the surrounding rock prevention strategy. 2.The tunnel surrounding rock risk monitoring system based on multi-source data fusion according to claim 1, characterized in that, The method comprises the following steps: collecting multi-dimensional construction data at monitoring device points and capturing time, acquiring a tunnel construction monitoring scheme, determining monitoring device points according to the tunnel construction monitoring scheme, collecting multi-dimensional construction data based on the monitoring device points, and recording corresponding data capturing time. 3.The tunnel surrounding rock risk monitoring system based on multi-source data fusion according to claim 1, characterized in that, The multi-dimensional construction data comprises face surrounding rock deformation data, support deformation data and comprehensive geological conditions; the monitoring device points are information collection device placing points determined according to the tunnel construction monitoring scheme.

4. A carrier device, characterized by The application further discloses a tunnel surrounding rock risk monitoring system based on multi-source data fusion, which comprises a memory and a processor; the memory is connected with the processor, and the memory stores the tunnel surrounding rock risk monitoring system based on multi-source data fusion.

Images