Tunnel blasting support safety monitoring method and system
By constructing a multi-dimensional evaluation index system and parallel factor analysis, combined with clustering methods, the problem of inaccurate assessment in tunnel blasting support safety monitoring was solved, and differentiated assessment and real-time monitoring of support structures were realized.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-12-29
- Publication Date
- 2026-04-07
AI Technical Summary
Existing methods for monitoring the safety of tunnel blasting support suffer from several drawbacks. Single-indicator assessments are inaccurate and fail to fully reflect the true safety status of the support structure. Furthermore, multi-indicator assessments fail to fully explore data patterns and relationships, resulting in assessment results that lack specificity and accuracy.
An evaluation index system was established, monitoring point data was collected, a support safety evaluation dataset was constructed, the contribution of the indicators was determined through parallel factor analysis, monitoring point clusters were formed by clustering, and the weights of the indicators were determined to achieve differentiated evaluation.
It improves the accuracy and relevance of safety assessments for tunnel support structures, enabling real-time monitoring and early warning, and adapting to safety changes in different support areas.
Smart Images
Figure CN121803302A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of tunnel safety monitoring technology, and in particular to a method and system for monitoring the safety of tunnel blasting support. Background Technology
[0002] In the field of tunnel engineering construction, tunnel support structures serve as a crucial line of defense for tunnel stability and construction safety. Their safety status directly impacts the smooth progress of the entire project and the safety of construction workers' lives and property. Therefore, real-time, accurate, and comprehensive monitoring and assessment of the safety of support structures during tunnel blasting is an issue that cannot be ignored.
[0003] Currently, while some methods have been applied in the safety monitoring of tunnel blasting support, most have significant shortcomings. On the one hand, some methods use only a single indicator to assess the safety of the support structure. This simplistic approach cannot comprehensively reflect the true safety status of the support structure under complex blasting environments, easily overlooking potential safety hazards and leading to inaccurate and unreliable assessment results. On the other hand, even when using multi-indicator comprehensive evaluation, a uniform fixed weight is often set for the entire tunnel cross-section or area, failing to fully explore the hidden patterns and relationships in the data. This makes it difficult to comprehensively consider the safety of the support structure from multiple dimensions, failing to reflect the differences in the dominant risk factors at different locations within the tunnel. Consequently, the safety assessment of different support points lacks specificity, and the accuracy of the assessment needs improvement. Summary of the Invention
[0004] To address the shortcomings of existing technologies, this invention provides a method and system for monitoring the safety of tunnel blasting support.
[0005] To achieve the above objectives, in a first aspect, the present invention provides a method for monitoring the safety of tunnel blasting support. The method includes the following steps: establishing an evaluation index system for assessing the safety of tunnel support structures, collecting evaluation index data from various monitoring points in the tunnel, and constructing a support safety evaluation dataset; determining the contribution of each evaluation index to the safety of different monitoring points based on parallel factor analysis decomposition using the support points as cluster centers; dividing the monitoring points into multiple monitoring point clusters based on the safety contribution, and determining the importance weight of the evaluation index in each monitoring point cluster; and calculating the safety score of each monitoring point in each monitoring point cluster using the corresponding evaluation index and importance weight to assess the safety of each point. This method quantifies the influence of each index on safety in different support areas based on parallel factor analysis decomposition, and combines this with clustering of monitoring points to achieve differentiated assessment of the safety of different support areas, thereby improving the accuracy of blasting support monitoring.
[0006] Optionally, the monitoring points include supported points and unsupported discrete points.
[0007] Optionally, the evaluation indicators include at least transverse stress, circumferential stress, strain, vertical displacement, horizontal displacement, vibration frequency, vibration velocity, vibration amplitude, and rock strength.
[0008] Optionally, the step of establishing an evaluation index system for assessing the safety of tunnel support structures and collecting evaluation index data from various monitoring points in the tunnel to construct a support safety evaluation dataset includes the following steps: An evaluation index system for assessing the safety of tunnel support structures is established, a monitoring network is deployed inside the tunnel, and the evaluation indexes are collected in real time according to a preset sampling frequency. The evaluation index data is preprocessed by edge computing nodes, and the preprocessed evaluation index data is used in the cloud to construct a support safety evaluation dataset.
[0009] Optionally, determining the contribution of each evaluation index to the safety of different monitoring points based on parallel factor analysis decomposition using the support safety evaluation dataset includes the following steps: A three-dimensional data array dataset of the tunnel is constructed based on the aforementioned support safety evaluation dataset; Based on the three-dimensional data array dataset, parallel factor analysis was used to determine the safety contribution of each evaluation indicator at the monitoring point.
[0010] Optionally, constructing a three-dimensional data array dataset of the tunnel based on the support safety evaluation dataset includes the following steps: In the support safety evaluation dataset, multiple data sequences of the monitoring points are obtained using the sliding window method; Based on the data sequence, multiple three-dimensional data arrays containing the monitoring points, the evaluation index data, and time are constructed to build the three-dimensional data array dataset.
[0011] Optionally, determining the safety contribution of each evaluation indicator at the monitoring point using parallel factor analysis decomposition based on the three-dimensional data array dataset includes the following steps: Define the range of pattern numbers; For any number of modes within the range of modes, based on the three-dimensional data array dataset, the reconstruction error of the three-dimensional data array for different time ranges by parallel factor analysis decomposition is obtained, and then the optimal number of modes is determined. Based on the optimal number of modes and the latest three-dimensional data array, the location factor matrix and index factor matrix are obtained by parallel factor analysis decomposition. The location contribution of different modes at the monitoring point is determined based on the location factor matrix, and the index contribution of different evaluation indicators under each mode is determined based on the index factor matrix. The security contribution of each evaluation indicator is calculated using a weighted mixed approach, based on the location contribution and the indicator contribution.
[0012] Optionally, the step of dividing the monitoring points into multiple monitoring point clusters based on the safety contribution, using the support point as the cluster center, and determining the importance weight of the evaluation index in each monitoring point cluster, includes the following steps: Set a distance threshold for each of the support points, and determine the support area of the support point with the support point as the center and the distance threshold as the radius. Combine the spatial distance between the support point and the unsupported discrete point to initially divide the clusters. For the unsupported discrete points that are clustered into multiple clusters, clustering optimization is performed using the security contribution to obtain multiple monitoring point clusters; For any of the monitoring point clusters, the average contribution of the corresponding indicators is calculated using the security contribution of the evaluation indicators and then normalized to obtain the importance weight of the evaluation indicators.
[0013] Optionally, for the unsupported discrete points clustered into multiple clusters, clustering optimization is performed using a security contribution to obtain multiple monitoring point clusters, including the following steps: The unsupported discrete points that are clustered into multiple clusters are denoted as monitoring points to be optimized, and the cluster centers corresponding to the monitoring points to be optimized are denoted as optional support points; The similarity between the monitoring point to be optimized and the optional support point is calculated using the safety contribution of the monitoring point to be optimized and the optional support point. The monitoring points to be optimized are classified into the clusters of the optional support points with the highest similarity, and finally multiple monitoring point clusters are obtained.
[0014] Secondly, this invention provides a tunnel blasting support safety monitoring system, which includes: a data acquisition device, a data output device, a processor, and a storage device. The storage device includes a computer-readable storage medium storing a computer program. The computer program includes program instructions, which, when executed by the processor, cause the processor to implement the tunnel blasting support safety monitoring method provided by this invention.
[0015] In summary, the present invention has at least the following beneficial effects: 1. This method constructs a multi-dimensional evaluation index system. Compared with the blasting support safety monitoring method that only uses a single index, this method can comprehensively and holistically evaluate the safety of tunnel support structures and improve the accuracy of safety assessment.
[0016] 2. This method constructs a three-dimensional data array dataset of tunnels and uses parallel factor analysis to decompose and determine the contribution of each evaluation indicator to the safety of different monitoring points. It decomposes and analyzes the data from multiple dimensions such as time, space and indicators to mine hidden patterns and relationships in the data, thereby more accurately quantifying the impact of each indicator on safety in different areas and providing a more scientific basis for subsequent clustering and weight determination.
[0017] 3. This method uses the support point as the cluster center and clusters the monitoring points according to their spatial location and safety contribution, forming multiple monitoring point clusters. This clustering method combines the spatial location of the support point and the safety contribution of each indicator, which can more reasonably divide the monitoring area and overcome the limitations of assessing the safety of a single support structure, thus achieving the assessment of the safety of the entire support area.
[0018] 4. This method determines the importance weight of evaluation indicators for each monitoring point cluster, taking into account the relative importance differences of indicators in different support areas. It can more accurately identify the key indicators for safety evaluation of different support areas in the tunnel, and realize differentiated assessment of the safety of different support areas.
[0019] 5. This method calculates the safety score of each monitoring point cluster using corresponding evaluation indicators and importance weights. Compared with using a uniform standard and weight to score all areas, this personalized scoring method takes into account the differences between different support areas. It can provide more realistic safety assessment results based on the characteristics of different areas and the importance of indicators, and provide more targeted decision-making basis for tunnel support safety management.
[0020] 6. This method first determines the importance weight of the evaluation indicators in each monitoring point cluster, and then evaluates the safety of each monitoring point based on the real-time collected data. The evaluation stage does not require complex data processing. Combined with the edge-cloud collaboration mechanism, it can reflect the safety changes in the tunnel support area in a timely manner, and can also realize real-time monitoring and early warning of tunnel support safety during blasting.
[0021] 7. A system adapted to the method is provided, which can improve the practicality of the method and facilitate its promotion. Attached Figure Description
[0022] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of this application and should not be regarded as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.
[0023] Figure 1 This is a flowchart illustrating a method for monitoring the safety of tunnel blasting support according to an embodiment of the present invention. Figure 2 A schematic diagram of the framework of a tunnel blasting support safety monitoring system according to an embodiment of the present invention. Detailed Implementation
[0024] Specific embodiments of the present invention will now be described in detail. It should be noted that the embodiments described herein are for illustrative purposes only and are not intended to limit the invention. In the following description, numerous specific details are set forth in order to provide a thorough understanding of the invention. However, it will be apparent to those skilled in the art that these specific details are not necessary to practice the invention. In other instances, well-known circuits, software, or methods have not been specifically described to avoid obscuring the invention.
[0025] Throughout this specification, references to "an embodiment," "an embodiment," "an example," or "an example" mean that a particular feature, structure, or characteristic described in connection with that embodiment or example is included in at least one embodiment of the invention. Therefore, the phrases "in an embodiment," "in an embodiment," "an example," or "an example" appearing in various places throughout the specification do not necessarily refer to the same embodiment or example. Furthermore, specific features, structures, or characteristics can be combined in one or more embodiments or examples in any suitable combination and / or sub-combination. Moreover, those skilled in the art will understand that the illustrations provided herein are for illustrative purposes and are not necessarily drawn to scale.
[0026] It should be noted in advance that, in one alternative embodiment, apart from being described independently, the same symbols or letters appearing in all formulas have the same meaning and value.
[0027] In one optional embodiment, please refer to Figure 1 This invention provides a method for safety monitoring of tunnel blasting support, the method comprising the following steps: S1. Establish an evaluation index system for evaluating the safety of tunnel support structures, collect evaluation index data from various monitoring points in the tunnel, and then construct a support safety evaluation dataset.
[0028] Step S1 specifically includes the following steps: S11. Establish an evaluation index system for evaluating the safety of tunnel support structures, deploy a monitoring network inside the tunnel, and collect the evaluation indexes in real time according to a preset sampling frequency.
[0029] Specifically, in this embodiment, the tunnel support structure is a series of support and protection structural systems set up in tunnel engineering to ensure construction safety, maintain the stability of the surrounding rock, and ensure the long-term performance of the tunnel. These include anchor bolt support, steel frame support, shotcrete support, and concrete lining structures, which function promptly after tunnel excavation, forming a load-bearing structure together with the surrounding rock. The evaluation index system for assessing the safety of the tunnel support structure includes, but is not limited to, nine indicators: transverse stress, circumferential stress, strain, vertical displacement, horizontal displacement, vibration frequency, vibration velocity, vibration amplitude, and rock strength. Monitoring points include support points and unsupported discrete points. Support points are monitoring points located on the support structure, which can be located inside or on the surface of the support structure; they can also be located at the junction of the support structure and the tunnel. Multiple support points can be set on the same support structure. Unsupported discrete points refer to points located near the support structure, not in direct contact with the support structure, and on the inner surface of the tunnel.
[0030] Furthermore, various sensors are deployed at the monitoring points to acquire different types of time-series data. Specifically, vibrating wire stress gauges are used to acquire transverse and circumferential stress data; resistance strain gauges are used to acquire strain data; laser displacement sensors are used to acquire vertical and horizontal displacement data; wireless vibration sensors are used to acquire vibration frequency data; velocity sensors are used to acquire velocity data for vibration analysis; and acceleration sensors are used to acquire vibration amplitude data. The types of sensors mentioned above can be selected according to actual needs. In other optional embodiments, other types of sensors can also be used to collect the above evaluation indicators in real time.
[0031] S12. The evaluation index data is preprocessed through edge computing nodes, and the preprocessed evaluation index data is used in the cloud to construct a support safety evaluation dataset.
[0032] Specifically, in this embodiment, the evaluation index data refers to the time series data of each evaluation index. Data collected by each sensor is transmitted to edge computing nodes via the Internet of Things (IoT). The edge computing nodes first timestamp each data point using their own clock module, and then preprocess the collected data. The preprocessing operations specifically include: identifying and removing outliers in the data using the quartile interval method; filling in missing values using the spline interpolation method; downsampling various time series data based on the timestamps using the direct extraction method to align the data of each evaluation index in time; and normalizing the evaluation index data using the maximum-minimum normalization method.
[0033] After the above preprocessing operations, the edge computing nodes transmit the time-series data of each evaluation indicator to the cloud via the Internet of Things. At this point, the cloud can uniformly store all evaluation indicator data, forming a support safety evaluation dataset. It should be noted that the support safety evaluation dataset adopts a structured storage method, and the data is organized and stored according to a multi-dimensional structure of "monitoring point number - evaluation indicator type - timestamp".
[0034] S2. Based on the support safety evaluation dataset, determine the contribution of each evaluation index to the safety of different monitoring points using parallel factor analysis decomposition.
[0035] Step S2 specifically includes the following steps: S21. Construct a three-dimensional data array dataset of the tunnel based on the support safety evaluation dataset.
[0036] Specifically, step S21 includes the following steps: S211. In the support safety evaluation dataset, multiple data sequences of the monitoring points are obtained using the sliding window method.
[0037] Specifically, in this embodiment, for the time series data of any evaluation indicator in the support safety evaluation dataset, the sliding window method is used to truncate it into multiple data sequences. The window length and sliding step size are both set to 5 days.
[0038] S212. Based on the data sequence, construct multiple three-dimensional data arrays containing the monitoring points, the evaluation index data, and time, to build the three-dimensional data array dataset.
[0039] Specifically, in this embodiment, data sequences of different evaluation indicators within the same time period are treated as a set of data, and this set of data is used to construct a three-dimensional tensor containing monitoring points, evaluation indicator data, and time, i.e., a three-dimensional data array. Furthermore, multiple three-dimensional data arrays can be obtained to construct a three-dimensional data array dataset. Within the three-dimensional data array dataset, the three-dimensional data arrays are stored in chronological order.
[0040] S22. Based on the three-dimensional data array dataset, use parallel factor analysis to determine the safety contribution of each evaluation indicator at the monitoring point.
[0041] Specifically, step S22 includes the following steps: S221. Define the range of pattern numbers.
[0042] Specifically, in this embodiment, determining the number of potential modes directly affects the accuracy and complexity of the Parallel Factor Analysis (PARAFAC) decomposition. For tunnels, potential risk modes (hereinafter referred to as modes) may include blasting impact, water seepage, and surface vibration. However, based on clearly defined evaluation indicators, when evaluating the support safety of the tunnel during blasting, it is not necessary to specify which potential modes exist; it is only necessary to determine the optimal number of modes to determine the contribution of different evaluation indicators to the safety of different areas. Before obtaining the optimal number of modes, a possible range of modes can be estimated based on expert experience. This range can be set conservatively, such as [3, 9]. However, it should be noted that the number of modes is a positive integer within the range of modes.
[0043] S222. For any number of modes within the range of modes, based on the three-dimensional data array dataset, obtain the reconstruction error of the three-dimensional data array for different time ranges by parallel factor analysis decomposition, and then determine the optimal number of modes.
[0044] Specifically, in this embodiment, the first k-1 (k not less than 2) three-dimensional data arrays are concatenated along the time axis into a three-dimensional tensor containing data over a longer time range, and denoted as the long-time-series three-dimensional tensor. Multiple long-time-series three-dimensional tensors can be obtained through this operation. For each long-time-series three-dimensional tensor, its PARAFAC decomposition is solved using the alternating least squares method to obtain the corresponding three factor matrices A, B, and C. This process can be expressed as: in, R is a long-term three-dimensional tensor; R is the number of modes; Let be the vector in the r-th column of the location factor matrix A, representing the contribution of the r-th potential mode at each monitoring point; Let be the vector in the r-th column of the indicator factor matrix B, representing the importance of each evaluation indicator in the r-th potential pattern; Let be the vector in the r-th column of the time factor matrix C, representing the activation level of the r-th latent mode over time; It represents the outer product.
[0045] Then, the decomposition results are used to analyze the long-term three-dimensional tensor. Reconstruct the tensor to obtain the reconstructed tensor, and calculate the reconstruction error according to the following formula: Where Error represents the reconstruction error. To reconstruct the tensor, It is the Frobenius norm.
[0046] For any possible R, calculate the reconstruction error of each long-time 3D tensor, then calculate the average reconstruction error of all long-time 3D tensors, and take this average reconstruction error as the reconstruction error under that R, denoted as . Minimize reconstruction error The corresponding R is taken as the number of optimal modes, and denoted as . .
[0047] S223. Based on the optimal number of modes and the latest three-dimensional data array, use parallel factor analysis to decompose and obtain the location factor matrix and the index factor matrix.
[0048] Specifically, in this embodiment, after determining the optimal number of modes, the latest three-dimensional data array is decomposed using PARAFAC decomposition to obtain the corresponding location factor matrix and index factor matrix.
[0049] S224. Determine the location contribution of different modes at the monitoring point based on the location factor matrix, and determine the index contribution of different evaluation indicators under each mode based on the index factor matrix.
[0050] Specifically, in this embodiment, the location factor matrix mathematically encodes the intensity distribution of each mode in space (monitoring point), where the vector in the r-th column represents the contribution of the r-th mode at each monitoring point, and the i-th row represents the contribution of different modes at the i-th monitoring point. Therefore, the contribution of different modes at a monitoring point, i.e., the location contribution, can be determined based on the location factor matrix.
[0051] Similarly, the indicator factor matrix mathematically encodes the importance of each evaluation indicator to the model under each model. The vector in the r-th column represents the importance of each evaluation indicator in the r-th model, and the j-th row represents the importance of the j-th evaluation indicator in different models. Therefore, the importance of different evaluation indicators under each model, i.e., the indicator contribution, can be determined based on the indicator factor matrix.
[0052] S225. Using the location contribution and the indicator contribution, calculate the security contribution of each of the evaluation indicators in a weighted mixed manner.
[0053] Specifically, in this embodiment, at the same monitoring point, multiple evaluation indicators may simultaneously affect its safety. Within the same tunnel, a single monitoring point may be affected by multiple modes simultaneously. Therefore, to achieve a more accurate comprehensive assessment of support safety, a weighted mixed approach is used to calculate the safety contribution of each evaluation indicator. The safety contribution satisfies the following relationship: in, Let i be the security contribution vector of the i-th monitoring point. The contribution of the j-th evaluation index at the i-th monitoring point. , The contribution of the r-th pattern to the position at the i-th monitoring point.
[0054] This embodiment constructs a three-dimensional data array dataset of the tunnel and uses parallel factor analysis to decompose and determine the contribution of each evaluation indicator to the safety of different monitoring points. The data is decomposed and analyzed from multiple dimensions such as time, space and indicators to uncover hidden patterns and relationships in the data. This allows for a more accurate quantification of the impact of each indicator on safety in different areas, providing a more scientific basis for subsequent clustering and weight determination.
[0055] S3. Using the support point as the cluster center, the monitoring point is divided into multiple monitoring point clusters according to the safety contribution, and the importance weight of the evaluation index in each monitoring point cluster is determined.
[0056] Specifically, step S3 includes the following steps: S31. Set a distance threshold for each of the support points, and determine the support area of the support point with the support point as the center and the distance threshold as the radius. Combine the spatial distance between the support point and the unsupported discrete point to initially divide the clusters.
[0057] Specifically, in this embodiment, the distance thresholds corresponding to each support point need to be determined through on-site exploration, but it should be ensured that all unsupported discrete points are located within at least one support area, i.e., overlapping areas between different support areas are allowed. After determining the support areas, for any unsupported discrete point, the Euclidean distance between it and all support points is calculated. Then, the Euclidean distance and the distance thresholds corresponding to each support point determine which support area(s) the unsupported discrete point belongs to, and it is then assigned to the corresponding support area. Thus, all monitoring points are initially divided into multiple clusters.
[0058] S32. For the unsupported discrete points that are clustered into multiple clusters, cluster optimization is performed using security contribution to obtain multiple monitoring point clusters.
[0059] Specifically, step S32 includes the following steps: S321. The unsupported discrete points that are clustered into multiple clusters are denoted as monitoring points to be optimized, and the cluster centers corresponding to the monitoring points to be optimized are denoted as optional support points.
[0060] S322. Using the safety contribution of the monitoring point to be optimized and the optional support point, calculate the similarity between the monitoring point to be optimized and the optional support point.
[0061] Specifically, in this embodiment, the similarity between the monitoring point to be optimized and the optional support point can be calculated using the following formula: Where s represents the similarity score, and w represents the security contribution vector of the monitoring point to be optimized. This is a vector representing the safety contribution of the optional support points. High similarity indicates a greater correlation between the patterns and indicators affecting the safety of the monitoring point to be optimized and the optional support points, suggesting they should be grouped into the same cluster. Conversely, low similarity indicates less correlation and a greater likelihood that they should not be grouped into the same cluster.
[0062] S323. The monitoring points to be optimized are classified into the clusters of the optional support points with the highest similarity, and finally multiple monitoring point clusters are obtained.
[0063] Specifically, in this embodiment, if there are multiple support points on the support structure, a single support point may constitute a monitoring point cluster. This embodiment uses the support points as cluster centers, and clusters the monitoring points based on their spatial location and safety contribution, forming multiple monitoring point clusters. This clustering method combines the spatial location of the support points with the safety contribution of each indicator, and can overcome the limitations of assessing the safety of a single support structure, achieving an assessment of the safety of the entire support area.
[0064] S33. For any of the monitoring point clusters, calculate the average contribution of the corresponding indicators using the security contribution of the evaluation indicators and perform normalization processing to obtain the importance weight of the evaluation indicators.
[0065] Specifically, in this embodiment, for any cluster of monitoring points, the average contribution of each indicator is calculated using the following formula: in, Let N be the average contribution of the j-th evaluation index to the m-th monitoring point cluster, and N be the number of monitoring points in the m-th monitoring point cluster. The contribution of the nth monitoring point in the mth monitoring point cluster to the safety of the nth monitoring point.
[0066] After calculating the average contribution of each evaluation indicator in the monitoring point cluster, the average contribution is normalized using the maximum-minimum normalization method to obtain the importance weight of each evaluation indicator in the monitoring point cluster.
[0067] This embodiment determines the importance weight of evaluation indicators for each monitoring point cluster, taking into account the relative importance differences of indicators in different support areas. This enables more accurate identification of key indicators for safety evaluation of different support areas in the tunnel, and achieves differentiated assessment of the safety of different support areas.
[0068] S4. For each of the monitoring point clusters, use the corresponding evaluation indicators and importance weights to calculate the safety score of each monitoring point in the monitoring point cluster, so as to evaluate the safety of each point.
[0069] Specifically, in this embodiment, all monitoring points within a cluster of monitoring points are considered to be within the same support area. Therefore, when calculating the safety score of each monitoring point in the same cluster, the same set of importance weights is used for both supported and unsupported discrete points.
[0070] For a cluster of monitoring points, after collecting real-time evaluation index data from each monitoring point, the evaluation index data can be normalized first, and then the safety score of each monitoring point can be calculated directly using the following formula: in, The security score for the nth monitoring point in the mth monitoring point cluster. Let the importance weight of the j-th evaluation index be the n-th monitoring point in the m-th monitoring point cluster. It is the normalized value of the j-th evaluation index of the n-th monitoring point in the m-th monitoring point cluster.
[0071] Furthermore, a safety threshold is set. If the safety score of a monitoring point is greater than this threshold, the monitoring point is considered to be in a safe state; otherwise, support or reinforcement is required. This embodiment recommends setting the safety threshold to 85 points. Relevant personnel can adjust it according to actual needs, such as decreasing the safety threshold to improve the performance of early warning.
[0072] Furthermore, the average safety score can be calculated using the safety scores of each monitoring point in the monitoring point cluster, which can then be used to evaluate the overall safety of the corresponding support area.
[0073] This embodiment calculates the safety score of each monitoring point cluster using corresponding evaluation indicators and importance weights. It takes into account the differences between different support areas and can provide more realistic safety assessment results based on the characteristics of different support areas and the importance of indicators. This provides a more targeted decision-making basis for tunnel support safety management in the area. Furthermore, by evaluating the safety of support points and the unsupported discrete points in their corresponding support areas, it is possible not only to determine the safety of the support structure itself, but also the safety of other locations in its support area. This is beneficial for preventing arch collapse and improving construction safety.
[0074] Furthermore, this embodiment, based on the location factor matrix and index factor matrix, is already sufficient for real-time support safety evaluation in scenarios including blasting. In other optional embodiments, the specific mode can be further deduced by combining the time factor matrix, improving the accuracy and interpretability of the evaluation. For example, if the vibration amplitude contributes highly in the index factor matrix of a certain mode, and the time factor matrix shows that the mode is activated instantaneously during blasting, then it can be inferred that the mode corresponds to "blasting impact".
[0075] It should be noted that in some cases, the actions described in the specification can be performed in different orders and still achieve the desired results. In this embodiment, the order of steps is given only to make the embodiment clearer and easier to explain, and not to limit it.
[0076] In one optional embodiment, please refer to Figure 2 To improve the practicality of this method and facilitate its promotion, the present invention also provides a tunnel blasting support safety monitoring system. The tunnel blasting support safety monitoring system includes: a data acquisition device 1, a data output device 2, a processor 3, and a storage device 4. The storage device 4 includes a computer-readable storage medium storing a computer program. The computer program includes program instructions, which, when executed by the processor 3, cause the processor 3 to perform the contents described in steps S1 to S4.
[0077] In summary, this invention has at least the following beneficial effects: This method constructs a multi-dimensional evaluation index system. Compared to blasting support safety monitoring methods that only use a single index, this method can comprehensively and holistically evaluate the safety of tunnel support structures, improving the accuracy of safety assessments. This method constructs a three-dimensional data array dataset of the tunnel and uses parallel factor analysis to decompose and determine the contribution of each evaluation index to the safety of different monitoring points. It decomposes and analyzes the data from multiple dimensions such as time, space, and indicators, uncovering hidden patterns and relationships within the data, thereby more accurately quantifying the impact of each index on safety in different regions. This method provides a more scientific basis for subsequent clustering and weight determination. Using support points as cluster centers, it clusters monitoring points based on their spatial location and safety contribution, forming multiple clusters. This clustering method combines the spatial location of support points with the safety contribution of each indicator, enabling more reasonable division of monitoring areas and overcoming the limitations of assessing the safety of a single support structure, thus achieving an assessment of the safety of the entire support area. Furthermore, this method determines the importance weight of evaluation indicators for each monitoring point cluster, considering the relative importance differences of indicators within different support areas, enabling more accurate identification of different support areas in the tunnel. This method identifies key indicators for safety assessment of different support areas, enabling differentiated evaluation. For each monitoring point cluster, it calculates a safety score for each point using corresponding evaluation indicators and importance weights. Compared to scoring all areas with uniform standards and weights, this personalized scoring method considers the differences between support areas and can provide more realistic safety assessment results based on the characteristics and importance of indicators in different areas, offering more targeted decision-making support for tunnel support safety management. The method first determines the importance weights of evaluation indicators in each monitoring point cluster, and then, based on the actual situation… The data collected in real time is used to assess the safety of each monitoring point. The assessment phase does not require complex data processing. Combined with the edge-cloud collaboration mechanism, it can reflect the safety changes in the tunnel support area in a timely manner. It can also realize real-time monitoring and early warning of tunnel support safety during blasting. It provides a reference method for support safety evaluation under other conditions. This method is not only applicable to tunnel blasting support safety monitoring, but can also be adapted to support safety evaluation in other scenarios, such as slope support and foundation pit excavation support, by adjusting the evaluation indicators. It provides a system adapted to the method, which can improve the practicality of the method and facilitate its promotion.
[0078] 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 therein. Such 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, and they should all be covered within the scope of the claims and specification of the present invention.
Claims
1. A method for safety monitoring of tunnel blasting support, characterized in that, Includes the following steps: Establish an evaluation index system for evaluating the safety of tunnel support structures, collect evaluation index data from various monitoring points in the tunnel, and then construct a support safety evaluation dataset. Based on the support safety evaluation dataset, the contribution of each evaluation index to the safety of different monitoring points is determined by parallel factor analysis decomposition. Using the support point as the cluster center, the monitoring point is divided into multiple monitoring point clusters according to the safety contribution, and the importance weight of the evaluation index in each monitoring point cluster is determined. For each of the monitoring point clusters, the corresponding evaluation indicators and importance weights are used to calculate the safety score of each monitoring point in the monitoring point cluster in order to evaluate the safety of each point.
2. The method for monitoring the safety of tunnel blasting support according to claim 1, characterized in that: The monitoring points include supported points and unsupported discrete points.
3. The method for monitoring the safety of tunnel blasting support according to claim 1, characterized in that: The evaluation indicators include at least transverse stress, circumferential stress, strain, vertical displacement, horizontal displacement, vibration frequency, vibration velocity, vibration amplitude, and rock strength.
4. The method for safety monitoring of tunnel blasting support according to claim 1, characterized in that, The process of establishing an evaluation index system for assessing the safety of tunnel support structures, collecting evaluation index data from various monitoring points in the tunnel, and then constructing a support safety evaluation dataset includes the following steps: An evaluation index system for assessing the safety of tunnel support structures is established, a monitoring network is deployed within the tunnel, and the evaluation indexes are collected in real time according to a preset sampling frequency. The evaluation index data is preprocessed by edge computing nodes, and the preprocessed evaluation index data is used in the cloud to construct a support safety evaluation dataset.
5. The method for safety monitoring of tunnel blasting support according to claim 1, characterized in that, The step of determining the contribution of each evaluation index to the safety of different monitoring points based on the support safety evaluation dataset and parallel factor analysis decomposition includes the following steps: A three-dimensional data array dataset of the tunnel is constructed based on the aforementioned support safety evaluation dataset; Based on the three-dimensional data array dataset, parallel factor analysis was used to determine the safety contribution of each evaluation indicator at the monitoring point.
6. The method for safety monitoring of tunnel blasting support according to claim 5, characterized in that, The step of constructing a three-dimensional data array dataset for the tunnel based on the support safety evaluation dataset includes the following steps: In the support safety evaluation dataset, multiple data sequences of the monitoring points are obtained using the sliding window method; Based on the data sequence, multiple three-dimensional data arrays containing the monitoring points, the evaluation index data, and time are constructed to build the three-dimensional data array dataset.
7. The method for monitoring the safety of tunnel blasting support according to claim 6, characterized in that, The step of determining the safety contribution of each evaluation indicator at the monitoring point using parallel factor analysis decomposition based on the three-dimensional data array dataset includes the following steps: Define the range of pattern numbers; For any number of modes within the range of modes, based on the three-dimensional data array dataset, the reconstruction error of the three-dimensional data array for different time ranges by parallel factor analysis decomposition is obtained, and then the optimal number of modes is determined. Based on the optimal number of modes and the latest three-dimensional data array, the location factor matrix and index factor matrix are obtained by parallel factor analysis decomposition. The location contribution of different modes at the monitoring point is determined based on the location factor matrix, and the index contribution of different evaluation indicators under each mode is determined based on the index factor matrix. The security contribution of each evaluation indicator is calculated using a weighted mixed approach, based on the location contribution and the indicator contribution.
8. The method for monitoring the safety of tunnel blasting support according to claim 2, characterized in that, The process of dividing the monitoring points into multiple clusters based on their safety contribution, using the support points as cluster centers, and determining the importance weight of the evaluation indicators in each cluster, includes the following steps: Set a distance threshold for each of the support points, and determine the support area of the support point with the support point as the center and the distance threshold as the radius. Combine the spatial distance between the support point and the unsupported discrete point to initially divide the clusters. For the unsupported discrete points that are clustered into multiple clusters, clustering optimization is performed using the security contribution to obtain multiple monitoring point clusters; For any of the monitoring point clusters, the average contribution of the corresponding indicators is calculated using the security contribution of the evaluation indicators and then normalized to obtain the importance weight of the evaluation indicators.
9. A method for monitoring the safety of tunnel blasting support according to claim 8, characterized in that, The step of clustering unsupported discrete points into multiple clusters using security contribution to optimize clustering and obtain multiple monitoring point clusters includes the following steps: The unsupported discrete points that are clustered into multiple clusters are denoted as monitoring points to be optimized, and the cluster centers corresponding to the monitoring points to be optimized are denoted as optional support points; The similarity between the monitoring point to be optimized and the optional support point is calculated using the safety contribution of the monitoring point to be optimized and the optional support point. The monitoring points to be optimized are classified into the clusters of the optional support points with the highest similarity, and finally multiple monitoring point clusters are obtained.
10. A tunnel blasting support safety monitoring system, characterized in that, The tunnel blasting support safety monitoring system includes: a data acquisition device, a data output device, a processor, and a storage device. The storage device includes a computer-readable storage medium storing a computer program. The computer program includes program instructions, which, when executed by the processor, cause the processor to implement the tunnel blasting support safety monitoring method as described in any one of claims 1-9.