Multi-source time series data fusion grading early warning method for soft rock stratum slope in rainy environment
By using a multi-source time-series data fusion and hierarchical early warning method for slopes with weak rock strata, we have achieved refined risk identification and dynamic early warning for slopes with weak rock strata. This solves the problem of inaccurate early warning level classification in traditional methods and improves the accuracy of early warning.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- ZHEJIANG RONGCHENG CONSTR ENG CO LTD
- Filing Date
- 2026-03-31
- Publication Date
- 2026-05-15
AI Technical Summary
Traditional slope safety monitoring methods fail to effectively consider the heterogeneity and spatiotemporal variability of weak rock strata in rainy environments, resulting in insufficient accuracy in early warning level classification and potential false alarms or missed alarms.
A multi-source time-series data fusion hierarchical early warning method is adopted. By fusing multi-source static data of various geometries in weak rock slopes, a first clustering process is performed to obtain sub-regions, and a second clustering process is performed on the multi-source time-series data of monitoring points. The early warning values are fused by combining static and dynamic values to achieve hierarchical early warning for slopes.
It improves the accuracy of early warning for slopes with weak rock strata, and realizes the accuracy of refined risk identification and dynamic early warning for slopes with weak rock strata.
Smart Images
Figure CN122050113A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of geotechnical engineering technology, and in particular to a method for hierarchical early warning of weak rock slopes in rainy environments by fusing multi-source time-series data. Background Technology
[0002] Weak rock strata typically refer to rock masses that are low in strength, easily weathered, and easily softened by water, such as mudstone, shale, and strongly weathered sandstone. Slopes composed of weak rock strata exhibit significant heterogeneity and spatiotemporal variability. In areas with frequent rainfall, the stability of slopes composed of weak rock strata is particularly problematic.
[0003] Traditional slope safety monitoring relies heavily on various sensors to acquire time-series data such as displacement, stress, seepage pressure, and rainfall, and issues warnings through single thresholds or simple models. This method often treats the slope as a whole or uses a uniform grid for partitioning, neglecting the heterogeneity and spatiotemporal variability of weak rock slopes. This leads to insufficient accuracy in classifying warning levels for weak rock slopes, potentially resulting in false alarms or missed alarms. Therefore, a superior multi-source time-series data fusion and hierarchical early warning scheme for weak rock slopes in rainy environments is needed to overcome these difficulties. Summary of the Invention
[0004] This application provides a multi-source time-series data fusion and hierarchical early warning method for soft rock slopes in rainy environments, in order to improve the safety and stability requirements of engineering construction.
[0005] In a first aspect, this application provides a method for hierarchical early warning of weak rock slopes under rainy conditions by fusing multi-source time-series data. The method includes: Multi-source static data of various geological points in the weak rock slope are fused to obtain static values of the sub-regions. Based on the static values, the geological points are subjected to a first clustering process to obtain multiple sub-regions in the weak rock slope. The multi-source static data is used to indicate the inherent attribute information of the geological points. The multi-source time-series data of each monitoring point in the sub-region are fused to obtain the dynamic value of the monitoring point. Based on the dynamic value and coordinate data, a second clustering operation is performed on the monitoring point to obtain multiple clusters. The multi-source time-series data is used to indicate the hydrological information of the monitoring point. The static values of the sub-region and the dynamic values of each cluster in the sub-region are fused to obtain the warning value of the sub-region, and the sub-region is given a graded warning based on the warning value.
[0006] Furthermore, the multi-source static data of various mass points in the weak rock slope are fused to obtain the static values of the sub-region, including: Based on preset categories, the multi-source static data of the geological points are classified into categories. The preset categories include at least soil and rock data, structural surface data, and slope geometric feature data. Based on the weights corresponding to different preset categories, the multi-source static data and location data of the geological point are fused to obtain the static value of the geological point.
[0007] Furthermore, the process of fusing the static values of the sub-region and the dynamic values of each cluster within the sub-region to obtain the warning value of the sub-region includes: The dynamic values of each cluster in the sub-region are weighted and averaged to obtain the mean dynamic value of the sub-region. The static value and the dynamic value of the sub-region are fused to obtain the warning value of the sub-region.
[0008] Furthermore, the step of weighted averaging of the dynamic values of each cluster in the sub-region to obtain the mean dynamic value of the sub-region includes: Multiple dynamic security levels are obtained, and the range of dynamic values corresponding to different dynamic security levels is different. Based on the dynamic values of the clusters, the dynamic security level to which the cluster belongs is determined. Based on the dynamic security level to which the cluster belongs, the weight of the cluster is determined, and based on the weight of the cluster, the dynamic value of each cluster in the sub-region is weighted and averaged to obtain the mean dynamic value of the sub-region.
[0009] Furthermore, determining the weight of the cluster based on the dynamic security level to which the cluster belongs includes: Obtain the first area covered by the sub-region and the second area covered by each cluster in the sub-region; The weights of the clusters to which the second area belongs are adjusted based on the ratio of the second area to the first area.
[0010] Furthermore, the step of providing tiered early warnings for the sub-regions based on the early warning value includes: Multiple warning levels are obtained, and the range of warning values corresponding to different warning levels is different; Based on the warning value of the sub-region, determine the warning level to which the sub-region belongs, and based on the warning information corresponding to the warning level to which the sub-region belongs, determine the warning information of the sub-region.
[0011] Secondly, this application provides a multi-source time-series data fusion and hierarchical early warning device for weak rock slopes in rainy environments. The multi-source time-series data fusion and hierarchical early warning device for weak rock slopes in rainy environments includes: The static module is used to fuse multi-source static data of geological points in the weak rock slope to obtain static values of the sub-regions, and based on the static values, to perform a first clustering process on the geological points to obtain multiple sub-regions in the weak rock slope. The multi-source static data is used to indicate the inherent attribute information of the geological points. The dynamic module is used to fuse the multi-source time-series data of each monitoring point in the sub-region to obtain the dynamic value of the monitoring point, and to perform a second clustering operation on the monitoring point based on the dynamic value and coordinate data to obtain multiple clusters. The multi-source time-series data is used to indicate the hydrological information of the monitoring point. The early warning module is used to fuse the static value of the sub-region and the dynamic value of each cluster in the sub-region to obtain the early warning value of the sub-region, and to perform hierarchical early warning for the sub-region based on the early warning value.
[0012] Thirdly, this application also provides a multi-source time-series data fusion hierarchical early warning system for soft rock slopes in rainy environments. The system includes a memory, a processor, and a computer program stored in the memory and running on the processor. When the processor executes the computer program, it implements the steps of any of the methods described above.
[0013] Fourthly, this application provides a computer-readable storage medium storing computer program instructions thereon, which, when executed by a processor, implement the above-described method. The computer-readable storage medium may be volatile or non-volatile.
[0014] Fifthly, this application provides an electronic device, comprising: a processor; and a memory for storing processor-executable instructions; wherein the processor is configured to implement the above-described method when executing instructions stored in the memory.
[0015] Sixthly, this application provides a computer program product including computer-readable code, or a non-volatile computer-readable storage medium carrying computer-readable code, wherein when the computer-readable code is run in a processor of an electronic device, the processor in the electronic device performs the above-described method.
[0016] In the embodiments of this specification, a first clustering process is performed on the geological points of the weak rock slope to obtain multiple sub-regions. A second clustering process is then performed on the monitoring points within each sub-region to obtain multiple clusters. Furthermore, based on the fusion of the static values corresponding to the sub-regions and the dynamic values of the clusters within those sub-regions, a warning value suitable for graded early warning is obtained. This process, through sub-regions and clusters, achieves a two-level refined division of the weak rock slope, facilitating refined risk identification of weak rock slopes. By fusing multi-source static data and multi-source time-series data, the resulting warning value considers both the geological nature of the weak rock slope (i.e., static values) and real-time hydrological stress (i.e., dynamic values), significantly improving the accuracy of dynamic early warning. Attached Figure Description
[0017] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.
[0018] Figure 1 The example illustrates a flowchart of a hierarchical early warning method for multi-source time-series data fusion on weak rock slopes in rainy environments; Figure 2 The diagram above illustrates a structural schematic of a multi-source time-series data fusion and hierarchical early warning device for soft rock slopes in rainy environments.
[0019] The accompanying drawings illustrate specific embodiments of this application, which will be described in more detail below. These drawings and descriptions are not intended to limit the scope of the concept in any way, but rather to illustrate the concept of this application to those skilled in the art through reference to particular embodiments. Detailed Implementation
[0020] Various exemplary embodiments, features, and aspects of this disclosure will now be described in detail with reference to the accompanying drawings. The same reference numerals in the drawings denote elements that have the same or similar functions. Although various aspects of the embodiments are shown in the drawings, they are not necessarily drawn to scale unless specifically indicated otherwise.
[0021] The term “exemplary” as used herein means “serving as an example, embodiment, or illustration.” Any embodiment illustrated herein as “exemplary” is not necessarily to be construed as superior to or better than other embodiments.
[0022] In this document, the term "and / or" is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent three cases: A alone, A and B simultaneously, and B alone. Furthermore, the term "at least one" in this document means any combination of at least two of any one or more elements. For example, including at least one of A, B, and C can mean including any one or more elements selected from the set consisting of A, B, and C.
[0023] Furthermore, to better illustrate this disclosure, numerous specific details are set forth in the following detailed description. Those skilled in the art will understand that this disclosure can be practiced without certain specific details. In some instances, methods, means, components, and circuits well known to those skilled in the art have not been described in detail in order to highlight the main points of this disclosure.
[0024] Figure 1 The flowchart illustrates a method for hierarchical early warning of weak rock slopes under rainy conditions based on a specific embodiment of this disclosure, using multi-source time-series data fusion. This method can be applied to a multi-source time-series data fusion hierarchical early warning device for weak rock slopes under rainy conditions. This device can be a terminal device, a server, or other processing equipment. The terminal device can be a user equipment (UE), mobile device, user terminal, terminal, cellular phone, cordless phone, personal digital assistant (PDA), handheld device, computing device, vehicle-mounted device, wearable device, etc.
[0025] In some possible implementations, the multi-source time-series data fusion hierarchical early warning method for weak rock slopes in rainy environments can be implemented by a processor calling computer-readable instructions stored in memory.
[0026] like Figure 1 As shown, the multi-source time-series data fusion and hierarchical early warning method for weak rock slopes under rainy conditions may include: Step S11: The multi-source static data of various geological points in the weak rock slope are fused to obtain the static value of the sub-region, and based on the static value, the geological points are subjected to a first clustering process to obtain multiple sub-regions in the weak rock slope.
[0027] The multi-source static data is used to indicate the inherent attribute information of the geological points. This multi-source static data comprises various types of data obtained from data collection at multiple sampling points (i.e., geological points) in a weak rock slope. The sources of multi-source static data can be diverse. For example, it can be data such as cone tip resistance and standard penetration test blow count collected by a static cone penetration test vehicle, or data such as soil resistivity and soil wave velocity collected by a high-density electrical resistivity meter and ground-penetrating radar. This specification does not specifically limit the sources and categories of multi-source static data; they can be determined according to the actual situation.
[0028] After obtaining multi-source static data, a fusion process can be performed to obtain static values that integrate information from multiple static data sources. These static values can then be used to evaluate the stability of geological points. In one possible implementation, the fusion process of multi-source static data from various geological points in a weak rock slope to obtain static values for the sub-region includes: Based on preset categories, the multi-source static data of the geological points are classified. Based on the weights corresponding to different preset categories, the multi-source static data and location data of the geological point are fused to obtain the static value of the geological point.
[0029] The preset categories include at least geotechnical data, structural surface data, slope geometric feature data, and location data. After obtaining multi-source static data collected by different devices, the multi-source static data can be categorized to determine the preset category to which each type of multi-source static data belongs.
[0030] Specifically, geotechnical data can include lithology, weathering degree, distribution of weak interlayers, soil type, particle composition, etc.; structural surface data can include the attitude (dip, dip angle), spacing, continuity, roughness, and infill properties of fissures, faults, and bedding; slope geometric feature data can include slope height, slope angle, slope aspect, profile shape, step width, etc.
[0031] In one example, different weights can be assigned to multi-source static data of different preset categories. For example... ( (Weights of geotechnical data) ( (Weights of the structural surface data) ( (The weights are for the slope geometric feature data). Furthermore, based on these weights, multi-source static data of geological points can be fused to obtain static values for the geological points. This process improves the accuracy of the static values through weighting.
[0032] The fusion processing of multi-source static data can be achieved by mapping the multi-source static data to a high-dimensional feature space to obtain corresponding vectors, connecting different multi-source static data in a preset order, and obtaining a vector that is a static joint feature vector. Furthermore, the static joint feature vector can be input into a machine learning model (such as support vector machine (SVM), random forest, or gradient boosting tree (XGBoost)) to obtain static values.
[0033] After obtaining the static values, the weak rock slope can be divided into regions through the first clustering process of the static values. This allows for refined monitoring of the weak rock slope by collecting multi-source time-series data from the resulting sub-regions. The algorithms used for the first and second clustering processes in this specification can be ST-DBSCAN, ST-DBSCAN, etc. The area covered by each cluster in the first clustering process is the sub-region.
[0034] Step S12: The multi-source time-series data of each monitoring point in the sub-region are fused to obtain the dynamic value of the monitoring point, and based on the dynamic value, a second clustering operation is performed on the monitoring point to obtain multiple clusters.
[0035] The multi-source time-series data is used to indicate the hydrological information of the monitoring points. This multi-source time-series data comprises various types of data obtained from data collection at multiple sampling points (i.e., monitoring points) within a sub-region. Multi-source time-series data can include rainfall data, groundwater level data, soil volumetric moisture content / saturation time-series data, etc. This specification does not specifically limit the source and category of multi-source time-series data; these can be determined based on actual circumstances. The multi-source time-series data fusion and hierarchical early warning method for soft rock slopes in rainy environments described in this specification can be an early warning based on multi-source time-series data within a single time window (e.g., the current time window).
[0036] Similar to multi-source static data, the categories of multi-source static data can be determined, and different weights can be assigned to multi-source time-series data of different preset categories. Furthermore, based on these weights, the multi-source time-series data of monitoring points can be fused to obtain the dynamic values of the monitoring points. This process improves the accuracy of static values through weighting.
[0037] The fusion processing of multi-source time series data can be achieved by mapping the multi-source time series data to a high-dimensional feature space to obtain corresponding vectors. The different multi-source time series data are then connected in a preset order, and the resulting vector is the dynamic joint feature vector. Furthermore, the dynamic joint feature vector is input into a machine learning model (such as support vector machine (SVM), random forest, or gradient boosting tree (XGBoost)) to obtain dynamic values.
[0038] After obtaining the dynamic values of each cluster in the sub-region, the dynamic values of each cluster can be fused. In one possible implementation, fusing the static values of the sub-region and the dynamic values of each cluster in the sub-region to obtain the warning value of the sub-region includes: The dynamic values of each cluster in the sub-region are weighted and averaged to obtain the mean dynamic value of the sub-region. The static value and the dynamic value of the sub-region are fused to obtain the warning value of the sub-region.
[0039] Specifically, the dynamic values of each cluster in a sub-region can be merged either by directly averaging them to obtain the mean dynamic value of the sub-region, or by setting different weights for different clusters and performing a weighted average based on the weights to obtain the mean dynamic value of the sub-region.
[0040] Clearly, the larger the dynamic value, the greater the risk of geological disasters in the area covered by that cluster in the weak rock slope. Therefore, clusters with large dynamic values can be assigned higher weights. In one possible implementation, the weighted average of the dynamic values of each cluster in the sub-region to obtain the mean dynamic value of the sub-region includes: Multiple dynamic security levels are obtained, and the range of dynamic values corresponding to different dynamic security levels is different. Based on the dynamic values of the clusters, the dynamic security level to which the cluster belongs is determined. Based on the dynamic security level to which the cluster belongs, the weight of the cluster is determined, and based on the weight of the cluster, the dynamic value of each cluster in the sub-region is weighted and averaged to obtain the mean dynamic value of the sub-region.
[0041] Specifically, multiple dynamic security levels can be set based on experience, with different levels corresponding to different ranges of dynamic values and their weights. In one example, a higher dynamic security level (corresponding to a higher dynamic value) can be assigned a larger weight.
[0042] After obtaining the dynamic values of the clusters, the dynamic safety level of each cluster can be determined based on these values. Then, the weight corresponding to each cluster can be determined based on its dynamic safety level. Furthermore, the dynamic values of each cluster in a sub-region can be weighted and averaged based on these weights to obtain the mean dynamic value for that sub-region. This mean dynamic value takes into account that clusters with higher dynamic values cover areas with greater risk, thus improving the sensitivity of the early warning system.
[0043] Besides considering the impact of dynamic values on the probability of geological disasters, the size of the area covered by a cluster also affects the probability of geological disasters. Therefore, clusters with larger areas within a sub-region can be assigned greater weights. In one possible implementation, determining the weight of a cluster based on its dynamic safety level includes: Obtain the first area covered by the sub-region and the second area covered by each cluster in the sub-region; The weights of the clusters to which the second area belongs are adjusted based on the ratio of the second area to the first area.
[0044] Specifically, we can first determine the first area covered by the sub-region and the second area covered by each cluster in the sub-region. Then we can determine the ratio of the second area to the first area (i.e., the area proportion of the cluster in the sub-region). We can then adjust the weight of the clusters based on the ratio. This adjustment can be to increase the weight of the clusters with larger area proportions and decrease the weight of the clusters with smaller area proportions. After adjusting the weights, we can then perform a weighted average of the dynamic values of the clusters.
[0045] After obtaining the static and dynamic values of a sub-region, they can be fused to obtain the warning value for that sub-region. This fusion method can be a weighted summation, with the weights set empirically. After obtaining the warning value, the sub-region can be classified into different warning levels based on the magnitude of the warning value. In one possible implementation, classifying the sub-region into different warning levels based on the warning value includes: Multiple warning levels are obtained, and the range of warning values corresponding to different warning levels is different; Based on the warning value of the sub-region, determine the warning level to which the sub-region belongs, and based on the warning information corresponding to the warning level to which the sub-region belongs, determine the warning information of the sub-region.
[0046] Specifically, multiple warning levels can be set based on experience, with different warning levels corresponding to different ranges of warning values. In one example, a larger range of warning values can be set for higher warning levels. Each warning level also needs to have different warning information, which can include the warning level. In this way, after obtaining the warning values for a sub-region, graded warnings for soft rock slopes can be achieved by using the warning levels and warning information corresponding to the warning values of the sub-regions.
[0047] Furthermore, the warning information can include risk heatmaps for each sub-region. Different colors can be added to clusters with different dynamic values in the heatmap. For example, a darker color can be added to the area covered by clusters with high dynamic values, and a lighter color can be added to the area covered by clusters with low dynamic values.
[0048] In the embodiments of this specification, a first clustering process is performed on the geological points of the weak rock slope to obtain multiple sub-regions. A second clustering process is then performed on the monitoring points within each sub-region to obtain multiple clusters. Furthermore, based on the fusion of the static values corresponding to the sub-regions and the dynamic values of the clusters within those sub-regions, a warning value suitable for graded early warning is obtained. This process, through sub-regions and clusters, achieves a two-level refined division of the weak rock slope, facilitating refined risk identification of weak rock slopes. By fusing multi-source static data and multi-source time-series data, the resulting warning value considers both the geological nature of the weak rock slope (i.e., static values) and real-time hydrological stress (i.e., dynamic values), significantly improving the accuracy of dynamic early warning.
[0049] The present invention also provides a multi-source time-series data fusion and hierarchical early warning device for weak rock slopes in rainy environments. Figure 2 This diagram illustrates a block diagram of a multi-source time-series data fusion-based hierarchical early warning device for weak rock slopes in rainy environments, according to an embodiment of this specification. This device can be a terminal device, a server, or other processing equipment. The terminal device can be a user equipment (UE), mobile device, user terminal, terminal, cellular phone, cordless phone, personal digital assistant (PDA), handheld device, computing device, vehicle-mounted device, wearable device, etc.
[0050] In some possible implementations, the multi-source time-series data fusion hierarchical early warning device for weak rock slopes in rainy environments can be implemented by a processor calling computer-readable instructions stored in memory.
[0051] like Figure 2 As shown, the multi-source time-series data fusion and hierarchical early warning device 20 for soft rock slopes in rainy environments may include: The static module 21 is used to fuse the multi-source static data of various geological points in the weak rock slope to obtain the static value of the sub-region, and to perform a first clustering process on the geological points based on the static value to obtain multiple sub-regions in the weak rock slope. The multi-source static data is used to indicate the inherent attribute information of the geological points. The dynamic module 22 is used to fuse the multi-source time-series data of each monitoring point in the sub-region to obtain the dynamic value of the monitoring point, and to perform a second clustering operation on the monitoring point based on the dynamic value and coordinate data to obtain multiple clusters. The multi-source time-series data is used to indicate the hydrological information of the monitoring point. The early warning module 23 is used to fuse the static value of the sub-region and the dynamic value of each cluster in the sub-region to obtain the early warning value of the sub-region, and to perform hierarchical early warning for the sub-region based on the early warning value.
[0052] Thirdly, this application also provides a multi-source time-series data fusion hierarchical early warning system for soft rock slopes in rainy environments. The system includes a memory, a processor, and a computer program stored in the memory and running on the processor. When the processor executes the computer program, it implements the steps of any of the methods described above.
[0053] Fourthly, embodiments of this application also provide a computer-readable storage medium storing computer program instructions thereon, which, when executed by a processor, implement the above-described method. The computer-readable storage medium may be volatile or non-volatile.
[0054] Fifthly, embodiments of this application also provide an electronic device, including: a processor; a memory for storing processor-executable instructions; wherein the processor is configured to implement the above method when executing instructions stored in the memory.
[0055] Sixthly, this application provides a computer program product including computer-readable code, or a non-volatile computer-readable storage medium carrying computer-readable code, wherein when the computer-readable code is run in a processor of an electronic device, the processor in the electronic device performs the above-described method.
[0056] This invention is now complete.
[0057] In summary, the first clustering process for geological points on weak rock slopes yields multiple sub-regions. A second clustering process is then performed on monitoring points within these sub-regions, resulting in multiple clusters. Furthermore, by fusing the static values of the sub-regions with the dynamic values of the clusters within those sub-regions, a tiered warning value can be obtained. This process, through sub-regions and clusters, achieves a two-level refined classification of weak rock slopes, facilitating more precise risk identification. The fusion of multi-source static and multi-source time-series data results in warning values that consider both the geological characteristics of the weak rock slopes (i.e., static values) and real-time hydrological stress (i.e., dynamic values), significantly improving the accuracy of dynamic warnings.
[0058] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the principles of the present invention should be included within the protection scope of the present invention.
Claims
1. A hierarchical early warning method for slopes with weak rock strata under rainy conditions, characterized in that, The method includes: Multi-source static data of various geological points in the weak rock slope are fused to obtain static values of the sub-regions. Based on the static values, the geological points are subjected to a first clustering process to obtain multiple sub-regions in the weak rock slope. The multi-source static data is used to indicate the inherent attribute information of the geological points. The multi-source time-series data of each monitoring point in the sub-region are fused to obtain the dynamic value of the monitoring point. Based on the dynamic value and coordinate data, a second clustering operation is performed on the monitoring point to obtain multiple clusters. The multi-source time-series data is used to indicate the hydrological information of the monitoring point. The static values of the sub-region and the dynamic values of each cluster in the sub-region are fused to obtain the warning value of the sub-region, and the sub-region is given a graded warning based on the warning value.
2. The method according to claim 1, characterized in that, The multi-source static data of various mass points in the weak rock slope are fused to obtain the static values of the sub-region, including: Based on preset categories, the multi-source static data of the geological points are classified into categories. The preset categories include at least soil and rock data, structural surface data, and slope geometric feature data. Based on the weights corresponding to different preset categories, the multi-source static data and location data of the geological point are fused to obtain the static value of the geological point.
3. The method according to claim 1, characterized in that, The process of fusing the static values of the sub-region and the dynamic values of each cluster in the sub-region to obtain the warning value of the sub-region includes: The dynamic values of each cluster in the sub-region are weighted and averaged to obtain the mean dynamic value of the sub-region. The static value and the dynamic value of the sub-region are fused to obtain the warning value of the sub-region.
4. The method according to claim 3, characterized in that, The step of weighted averaging of the dynamic values of each cluster in the sub-region to obtain the mean dynamic value of the sub-region includes: Multiple dynamic security levels are obtained, and the range of dynamic values corresponding to different dynamic security levels is different. Based on the dynamic values of the clusters, the dynamic security level to which the cluster belongs is determined. Based on the dynamic security level to which the cluster belongs, the weight of the cluster is determined, and based on the weight of the cluster, the dynamic value of each cluster in the sub-region is weighted and averaged to obtain the mean dynamic value of the sub-region.
5. The method according to claim 4, characterized in that, The step of determining the weight of the cluster based on the dynamic security level to which the cluster belongs includes: Obtain the first area covered by the sub-region and the second area covered by each cluster in the sub-region; The weights of the clusters to which the second area belongs are adjusted based on the ratio of the second area to the first area.
6. The method according to claim 3, characterized in that, The step of providing tiered early warnings for the sub-regions based on the early warning value includes: Multiple warning levels are obtained, and the range of warning values corresponding to different warning levels is different; Based on the warning value of the sub-region, determine the warning level to which the sub-region belongs, and based on the warning information corresponding to the warning level to which the sub-region belongs, determine the warning information of the sub-region.
7. A device for hierarchical early warning of weak rock slopes through multi-source time-series data fusion in rainy environments, characterized in that, include: The static module is used to fuse multi-source static data of geological points in the weak rock slope to obtain static values of the sub-regions, and based on the static values, to perform a first clustering process on the geological points to obtain multiple sub-regions in the weak rock slope. The multi-source static data is used to indicate the inherent attribute information of the geological points. The dynamic module is used to fuse the multi-source time-series data of each monitoring point in the sub-region to obtain the dynamic value of the monitoring point, and to perform a second clustering operation on the monitoring point based on the dynamic value and coordinate data to obtain multiple clusters. The multi-source time-series data is used to indicate the hydrological information of the monitoring point. The early warning module is used to fuse the static value of the sub-region and the dynamic value of each cluster in the sub-region to obtain the early warning value of the sub-region, and to perform hierarchical early warning for the sub-region based on the early warning value.
8. An electronic device, characterized in that, include: processor; as well as A memory configured to store computer-executable instructions configured to be executed by the processor, the executable instructions including steps for performing the method as described in any one of claims 1 to 6.
9. A storage medium, characterized in that, The storage medium is used to store computer-executable instructions that cause the computer to perform the method as described in any one of claims 1 to 6.
10. A computer program product, characterized in that, Includes a computer program that, when executed by a processor, implements the method described in any one of claims 1 to 6.