Evaluation method and system for vegetation restoration of bedding rock slope

Abstract

The invention relates to the technical field of ecological restoration evaluation, and discloses a vegetation restoration evaluation method and system for a bedding rock slope, and the method comprises the steps: collecting the vegetation information of a to-be-evaluated slope, and determining a restoration evaluation value; collecting humidity data at a plurality of positions in a preset time period to establish a time-humidity data chain, and constructing a humidity data set and a representative data chain; comparing with a historical humidity data set to determine predicted humidity data; and comparing the predicted humidity data with a humidity threshold value, judging whether to adjust the restoration evaluation value or not according to a comparison result, and when the restoration evaluation value is judged to be adjusted, determining a slope risk coefficient according to the predicted humidity data, and adjusting the restoration evaluation value according to the slope risk coefficient. Compared with a traditional static assessment method, the method can sense environment changes in real time, early warns the risk of slope stability reduction in advance, avoids hidden dangers caused by restoration effect overestimation, and improves the safety and precision of slope management.

Description

Technical Field

[0001] The present invention relates to the technical field of ecological restoration assessment, and in particular, to a method and system for evaluating vegetation restoration of bedding rock slopes. Background Art

[0002] Bedding rock slopes are widely distributed in mountainous areas and infrastructure construction regions. Due to the fact that their structural planes are in the same direction as the slope surface, they are extremely prone to bedding sliding under the action of external loads, precipitation, and weathering, which affects the slope stability. To reduce the risk of slope landslides, vegetation restoration has become a common and sustainable engineering measure. Vegetation can not only improve the microenvironment of the slope surface, reduce surface erosion, but also enhance the shear strength of the rock and soil mass through the root reinforcement effect, improving the slope stability. Therefore, the method for evaluating vegetation restoration of bedding rock slopes has important engineering significance.

[0003] Traditional methods for evaluating vegetation restoration mainly rely on green plant coverage and root distribution. Usually, means such as unmanned aerial vehicle remote sensing, ground surveys, or remote sensing image analysis are used to monitor the vegetation conditions of slopes, and the restoration effects are calculated based on empirical models. However, the long-term impact of humidity on slope stability is not fully considered. Changes in precipitation and soil humidity may lead to a decrease in the shear strength of the structural plane, thus affecting the slope safety. Currently, it is impossible to adjust the restoration evaluation value according to the humidity change trend. For example, when it is predicted that the future increase in humidity may lead to a decrease in the slope safety factor, the traditional method cannot timely correct the restoration evaluation value, which may overestimate the restoration effect and increase the risk of slope instability.

[0004] Therefore, it is necessary to design a method for evaluating vegetation restoration of bedding rock slopes to solve the problems existing in the current technology. Summary of the Invention

[0005] In view of this, the present invention proposes a method for evaluating vegetation restoration of bedding rock slopes, aiming to solve the problem that the current lack of dynamic assessment of the influence of humidity results in low reliability of vegetation restoration evaluation results.

[0006] On the one hand, the present invention proposes a method for evaluating vegetation restoration of bedding rock slopes, including: Collecting the vegetation information of the slope to be evaluated, where the vegetation information includes the green plant area and the average root depth, and preliminarily determining the restoration evaluation value according to the vegetation information; Based on humidity sensors, collecting humidity data at several locations within a preset time period of the slope to be evaluated, establishing a time-humidity data chain for the data of each humidity sensor, constructing a humidity data set from all the time-humidity data chains, and determining the representative humidity data of each moment of the slope to be evaluated based on Gaussian mixture distribution and constructing a representative data chain; Compare the representative data chain with the historical humidity data set, and determine the predicted humidity data according to the comparison result. The historical humidity data set includes a number of historical representative data chains and a number of historical verified humidity data, and each historical representative data chain corresponds to a historical verified humidity data; Compare the predicted humidity data with the humidity threshold, and determine whether to adjust the repair evaluation value according to the comparison result. When it is determined to adjust the repair evaluation value, determine the slope risk coefficient according to the predicted humidity data, and adjust the repair evaluation value according to the slope risk coefficient.

[0007] Further, when initially determining the repair evaluation value according to the vegetation information, it includes: ; Among them, S represents the repair evaluation value, As represents the green plant area, Az represents the total slope area, ds represents the average root depth, dz represents the critical root planting depth, τs represents the measured root shear strength, and τz represents the reference root shear strength.

[0008] Further, when determining the representative humidity data of each moment of the slope to be evaluated based on the Gaussian mixture distribution and constructing the representative data chain, it includes: The expression of the kernel density function is:

[0009] Among them, n represents the total number of humidity data collected at the same moment, h represents the smoothing bandwidth, represents the i-th humidity data collected at the same moment, and x represents the value of a certain frequency point to be estimated; Take the humidity data corresponding to the highest frequency of the kernel density as the representative data at this moment; Combine the representative data of each moment to construct the representative data chain.

[0010] Further, when comparing the representative data chain with the historical humidity data set and determining the predicted humidity data according to the comparison result, it includes: When there is data in the historical humidity data set whose similarity with the representative data chain is greater than the similarity threshold, take the historical verified humidity data of the historical representative data chain with the maximum similarity as the predicted humidity data; When the similarity between all historical representative data chains in the historical humidity data set and the representative data chain is less than or equal to the similarity threshold, train the LSTM model based on the historical humidity data set, and determine the predicted humidity data according to the LSTM model.

[0011] Further, when determining the predicted humidity data according to the LSTM model, it includes: Window-divide each humidity data in the representative data chain according to the time series as input data, obtain the output result of the LSTM model, and use the output result as the predicted humidity data.

[0012] Further, when judging whether to adjust the repair evaluation value according to the comparison result, it includes: Compare the predicted humidity data with the humidity threshold. When the predicted humidity data is greater than the humidity threshold, it is determined to adjust the repair evaluation value; when the predicted humidity data is less than or equal to the humidity threshold, it is determined not to adjust the repair evaluation value.

[0013] Further, when determining the slope risk coefficient according to the predicted humidity data, it includes: Obtain the humidity difference according to the predicted humidity data and the humidity threshold. The humidity difference is the difference between the predicted humidity data and the humidity threshold. Compare the humidity difference with the first preset difference and the second preset difference respectively, and determine the cohesion according to the comparison result; the first preset difference is less than the second preset difference; When the humidity difference is less than or equal to the first preset difference, determine the cohesion as the first preset cohesion; when the humidity difference is greater than the first preset difference and less than the second preset difference, determine the cohesion as the second preset cohesion; when the humidity difference is greater than the second preset difference, determine the cohesion as the third preset cohesion; the first preset cohesion is greater than the second preset cohesion, the second preset cohesion is greater than the third preset cohesion, and the third preset cohesion is greater than or equal to 0.

[0014] Further, when determining the slope risk coefficient according to the predicted humidity data, it also includes:

[0015] Wherein, F represents the slope risk coefficient, n represents the cohesion, the height of the slope is H, the slope angle is α, the dip angle of the structural plane is θ, γ is the rock unit weight, and φ is the internal friction angle.

[0016] Further, when adjusting the repair evaluation value according to the slope risk coefficient, it includes: Compare the slope risk coefficient with the minimum slope risk coefficient. The minimum slope risk coefficient is 1, and determine the adjustment coefficient to adjust the repair evaluation value according to the comparison result; The adjustment coefficient is in a direct proportional relationship with the slope risk coefficient, and the value range of the adjustment coefficient is (0, 1). The adjusted repair evaluation value is the product of the repair evaluation value and the adjustment coefficient.

[0017] Compared with the prior art, the beneficial effects of the present invention are as follows: Through vegetation information collection, humidity dynamic monitoring, data chain modeling, and risk assessment adjustment analysis, the dynamic assessment of the vegetation restoration effect of bedding rock slopes is realized. The vegetation information of the slope is obtained by using unmanned aerial vehicle remote sensing or ground measurement, including the green plant coverage area and the average root depth to calculate the restoration evaluation value. Humidity sensors are used to collect humidity data at multiple positions on the slope for a long time and stored in the form of a time-humidity data chain to construct a comprehensive humidity data set, avoiding the problem of insufficient accuracy caused by scattered data points in the traditional method. The Gaussian mixture distribution model is used to statistically analyze the humidity data to extract the most representative humidity data and reduce the influence of single-point data fluctuations on the evaluation results. By comparing the current representative data chain with the historical humidity data set, the humidity change trend in the future period is predicted, making the evaluation method forward-looking. When the predicted humidity data exceeds the safety threshold, the slope risk coefficient is further calculated, and combined with the slope stability theory, based on the influence of humidity on the shear strength of the slope, the restoration evaluation value is dynamically adjusted to ensure the reliability of the evaluation results. Compared with the traditional static evaluation method, it can sense environmental changes in real time, early warn of the risk of reduced slope stability, avoid the hidden dangers brought by overestimating the restoration effect, and improve the safety and accuracy of slope management.

[0018] On the other hand, the present application also provides a vegetation restoration evaluation system for bedding rock slopes, which is applied to the above-mentioned vegetation restoration evaluation method for bedding rock slopes, and includes: A collection unit configured to collect the vegetation information of the slope to be evaluated, the vegetation information including the green plant area and the average root depth, and preliminarily determine the restoration evaluation value according to the vegetation information; A processing unit configured to collect humidity data at several locations on the slope to be evaluated within a preset period based on humidity sensors, establish a time-humidity data chain for the data of each humidity sensor, construct a humidity data set from all the time-humidity data chains, and determine the representative humidity data at each moment of the slope to be evaluated based on the Gaussian mixture distribution and construct a representative data chain; A judgment unit configured to compare the representative data chain with the historical humidity data set and determine the predicted humidity data according to the comparison result, the historical humidity data set including several historical representative data chains and several historical verification humidity data, and each historical representative data chain corresponding to a historical verification humidity data; An adjustment unit configured to compare the predicted humidity data with the humidity threshold, judge whether to adjust the restoration evaluation value according to the comparison result, and when it is determined to adjust the restoration evaluation value, determine the slope risk coefficient according to the predicted humidity data and adjust the restoration evaluation value according to the slope risk coefficient.

[0019] It is understandable that the above-mentioned method and system for evaluating the vegetation restoration of bedding rock slopes have the same beneficial effects, which will not be elaborated here. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] By reading the detailed description of the preferred embodiments below, various other advantages and benefits will become clear to those of ordinary skill in the art. The drawings are only for the purpose of showing the preferred embodiments and are not considered as a limitation of the present invention. Moreover, throughout the drawings, the same reference numerals are used to represent the same components. In the drawings: Figure 1 is a flowchart of a method for evaluating the vegetation restoration of bedding rock slopes provided by an embodiment of the present invention; Figure 2 is a parameter schematic diagram of the slope risk coefficient in the method for evaluating the vegetation restoration of bedding rock slopes provided by an embodiment of the present invention; Figure 3 is a functional block diagram of a system for evaluating the vegetation restoration of bedding rock slopes provided by an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0021] The exemplary embodiments of the present disclosure will be described in more detail below with reference to the drawings. Although the exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided so that the present disclosure can be more thoroughly understood and the scope of the present disclosure can be completely conveyed to those skilled in the art. It should be noted that, without conflict, the embodiments in the present invention and the features in the embodiments can be combined with each other. The present invention will be described in detail below with reference to the drawings and in combination with the embodiments.

[0022] In some embodiments of the present application, referring to Figure 1-2 as shown, a method for evaluating the vegetation restoration of bedding rock slopes includes: S100: Collect the vegetation information of the slope to be evaluated, where the vegetation information includes the green plant area and the average root depth, and preliminarily determine the restoration evaluation value according to the vegetation information.

[0023] S200: Based on the humidity sensors, collect the humidity data at several locations on the slope to be evaluated within a preset time period, establish a time-humidity data chain for each humidity sensor's data, construct a humidity data set from all the time-humidity data chains, and determine the representative humidity data at each moment of the slope to be evaluated based on the Gaussian mixture distribution and construct a representative data chain.

[0024] S300: Compare the representative data chain with the historical humidity data set, and determine the predicted humidity data according to the comparison result. The historical humidity data set includes a number of historical representative data chains and a number of historical verified humidity data, and each historical representative data chain corresponds to a historical verified humidity data.

[0025] S400: Compare the predicted humidity data with the humidity threshold, and judge whether to adjust the repair evaluation value according to the comparison result. When it is determined to adjust the repair evaluation value, determine the slope risk coefficient according to the predicted humidity data, and adjust the repair evaluation value according to the slope risk coefficient.

[0026] Specifically, in S100, through drone remote sensing, ground measurement or remote sensing images, collect the vegetation information of the slope, including the green plant coverage area (used to measure the degree of vegetation restoration) and the average root depth (used to evaluate the reinforcement effect of vegetation on the rock and soil structural surface). Based on these data, initially determine the repair evaluation value. In S200, considering the direct impact of soil humidity on slope stability, multiple humidity sensors are arranged in different areas of the slope to collect humidity data for a long time, and a time-humidity data chain is constructed. Traditional humidity monitoring methods usually use simple averages or single-point measurements, which are difficult to accurately reflect the spatio-temporal changes of slope humidity. The Gaussian mixture distribution model is used to statistically model the collected humidity data to remove outliers and extract the most representative humidity data to form a representative data chain. In S300, in order to further improve the scientific nature of the evaluation, a historical humidity data set is introduced, which includes a number of historical representative data chains (i.e., humidity change trends in different historical periods) and historical verified humidity data (humidity values predicted historically and verified by actual observations). By comparing the current representative data chain with historical data, using pattern recognition or time series analysis methods, predict the humidity change trend in the future for a period of time to obtain the predicted humidity data. In S400, to ensure that the repair evaluation value can reflect the actual slope repair situation, compare the predicted humidity data with the set humidity threshold. When the predicted humidity exceeds the threshold, it will cause the shear strength of the slope to decrease, thereby affecting stability. A slope risk coefficient is introduced to quantify the impact of humidity changes on slope stability. According to the change of the risk coefficient, adjust the initially determined repair evaluation value to obtain a final repair evaluation value that is more in line with the actual stability.

[0027] It is understandable that by integrating vegetation information collection, humidity data chain modeling, historical comparison prediction, and risk adjustment, the optimization of vegetation restoration assessment based on dynamic humidity changes is achieved. Compared with the traditional static assessment method that only relies on green plant coverage and root depth, it can perceive humidity changes in real time, predict potential risks in advance, and dynamically adjust the restoration assessment value, thereby improving the accuracy and reliability of the assessment. The Gaussian mixture distribution is used to extract representative humidity data, effectively avoiding the calculation deviation caused by excessive humidity fluctuations or high data dispersion in the traditional method. At the same time, through historical data comparison and slope risk coefficient correction, it is ensured that the final assessment result can accurately reflect the long-term stability of the slope and the vegetation restoration effect.

[0028] In some embodiments of the present application, when initially determining the restoration assessment value according to vegetation information, it includes: ; Where S represents the restoration assessment value, As represents the green plant area, Az represents the total area of the slope, ds represents the average root depth, dz represents the critical root depth, τs represents the measured root shear strength, and τz represents the reference root shear strength.

[0029] It is understandable that the measured average root depth can be determined by sampling analysis or radar detection. The critical root depth, that is, the depth at which the plant roots have a significant effect on the slope stability, can set an empirical value based on the soil type and vegetation species. The measured root shear strength can be determined by in-situ shear tests. The reference root shear strength usually selects the root shear strength of the dominant vegetation in this area as the standard value. Compared with the assessment method based only on visual remote sensing data, it not only considers the surface distribution of vegetation (green plant area), but also quantifies the depth influence and shear performance of vegetation roots, ensuring that the assessment value can truly reflect the contribution of vegetation to slope stability.

[0030] In some embodiments of the present application, when determining the representative humidity data of each moment of the slope to be evaluated based on the Gaussian mixture distribution and constructing a representative data chain, it includes: The expression of the kernel density function is:

[0031] Where n represents the total number of humidity data collected at the same moment, h represents the smoothing bandwidth, represents the i-th humidity data collected at the same moment, and x represents the value of a certain frequency point to be estimated.

[0032] The humidity data corresponding to the highest frequency of the kernel density is used as the representative data at this moment.

[0033] The representative data of each moment is combined to construct a representative data chain.

[0034] Specifically, the smoothing bandwidth is calculated as follows:

[0035] where h represents the smoothing bandwidth, β1 represents the data skewness, β2 represents the data kurtosis, and k represents the correction coefficient.

[0036] Among them, the data skewness is calculated as follows:

[0037] The data kurtosis is calculated as follows: ; where represents the mean of the humidity data collected at the same moment, and σ represents the standard deviation of the humidity data collected at the same moment.

[0038] It can be understood that by combining the Gaussian mixture distribution with kernel density estimation, the most representative humidity value is extracted from the data of multiple humidity sensors, avoiding the distortion problem that may be caused by the traditional mean calculation method. At the same time, an adaptive bandwidth calculation method is adopted to dynamically adjust the smoothing bandwidth based on the data skewness and kurtosis, improving the accuracy of kernel density estimation and making the humidity representative data more in line with the actual situation of the slope. This solution constructs a representative humidity data chain for the time series, providing high-quality data input for subsequent humidity trend prediction and dynamic adjustment of the repair evaluation value, making the evaluation more accurate, and further improving the reliability and stability of the slope vegetation restoration effect.

[0039] In some embodiments of the present application, when comparing the representative data chain with the historical humidity data set and determining the predicted humidity data according to the comparison result, it includes: when there is data in the historical humidity data set whose similarity to the representative data chain is greater than the similarity threshold, the historical verification humidity number of the historical representative data chain corresponding to the maximum similarity is used as the predicted humidity data. When the similarity between all historical representative data chains in the historical humidity data set and the representative data chain is less than or equal to the similarity threshold, an LSTM model is trained based on the historical humidity data set, and the predicted humidity data is determined according to the LSTM model.

[0040] In some embodiments of the present application, when determining the predicted humidity data according to the LSTM model, it includes: dividing each humidity data in the representative data chain into windows according to the time series as input data, obtaining the output result of the LSTM model, and using the output result as the predicted humidity data.

[0041] Specifically, a forgetting gate of the LSTM model is pre-constructed: .

[0042] Input gate: 。

[0043] 。

[0044] Cell state update: 。

[0045] Output gate: 。

[0046] 。

[0047] Among them, represents the output of the forget gate, represents the weight matrix of the forget gate, and its value range is [-1, 1], represents the hidden state at the previous moment, represents the input data, represents the bias term of the forget gate, and σ is the sigmoid function, represents the output of the input gate, represents the candidate cell state, and its value range is (-1, 1), represents the weight matrix of the input gate, and its value range is [-1, 1], represents the weight matrix of the candidate cell state, and its value range is [-1, 1], 、 respectively represent the bias terms of the input gate and the candidate cell state, represents the cell state at the current moment, represents the cell state at the previous moment, represents the output of the output gate, represents the predicted data, represents the weight matrix of the output gate, and its value range is [-1, 1], represents the bias term of the output gate.

[0048] It can be understood that by combining similarity matching and LSTM deep learning, a humidity prediction method is constructed. Historical humidity data matching is preferentially used to ensure that the prediction results have high reliability and computational efficiency; when historical data is insufficient, the LSTM model is used for time series prediction, so that the prediction results have strong generalization ability and can adapt to complex slope environment changes. In addition, LSTM adopts the structure of forget gate, input gate, cell state update, and output gate, which solves the long-term dependence problem of humidity time series data, makes the prediction results more stable, and thus improves the scientificity and accuracy of slope repair evaluation.

[0049] In some embodiments of the present application, when determining whether to adjust the repair evaluation value according to the comparison result, it includes: comparing the predicted humidity data with the humidity threshold. When the predicted humidity data is greater than the humidity threshold, it is determined to adjust the repair evaluation value. When the predicted humidity data is less than or equal to the humidity threshold, it is determined not to adjust the repair evaluation value.

[0050] In some embodiments of the present application, when determining the slope risk coefficient according to the predicted humidity data, it includes: obtaining the humidity difference according to the predicted humidity data and the humidity threshold, where the humidity difference is the difference between the predicted humidity data and the humidity threshold, comparing the humidity difference with the first preset difference and the second preset difference respectively, and determining the cohesion according to the comparison result. The first preset difference is less than the second preset difference.

[0051] Specifically, when the humidity difference is less than or equal to the first preset difference, the cohesion is determined to be the first preset cohesion. When the humidity difference is greater than the first preset difference and less than the second preset difference, the cohesion is determined to be the second preset cohesion. When the humidity difference is greater than the second preset difference, the cohesion is determined to be the third preset cohesion. The first preset cohesion is greater than the second preset cohesion, the second preset cohesion is greater than the third preset cohesion, and the third preset cohesion is greater than or equal to 0.

[0052] In some embodiments of the present application, when determining the slope risk coefficient according to the predicted humidity data, it further includes:

[0053] Wherein, F represents the slope risk coefficient, n represents the cohesion, the height of the slope is H, the slope angle is α, the dip angle of the structural plane is θ, γ is the rock unit weight, and φ is the internal friction angle.

[0054] Specifically, in the figure, ACE is the structural plane. When the whole slope structural plane slides, the cohesion is 0, and the internal friction angle is equal to the dip angle of the structural plane. Therefore, the minimum slope risk coefficient is 1.

[0055] In some embodiments of the present application, when adjusting the repair evaluation value according to the slope risk coefficient, it includes: comparing the slope risk coefficient with the minimum slope risk coefficient, where the minimum slope risk coefficient is 1, and determining the adjustment coefficient according to the comparison result to adjust the repair evaluation value. The adjustment coefficient is in a direct proportion relationship with the slope risk coefficient, and the value range of the adjustment coefficient is (0, 1). The adjusted repair evaluation value is the product of the repair evaluation value and the adjustment coefficient.

[0056] It is understandable that by comparing the predicted humidity data with the threshold value, it is ensured that the adjustment of the repair evaluation value is based on real-time environmental data, improving the evaluation accuracy. By adjusting the cohesion according to the humidity change and further calculating the slope risk coefficient, the evaluation is more in line with the actual characteristics. Dynamically adjusting the repair evaluation value according to the slope risk coefficient enables the evaluation system to adapt to the slope stability under different humidity conditions and enhances the reliability of slope vegetation restoration. Controlling the repair evaluation value through the adjustment coefficient is beneficial to strengthening the repair efforts in high-risk areas and improving the overall repair effect.

[0057] In the above embodiment, through vegetation information collection, humidity dynamic monitoring, data chain modeling and risk assessment adjustment analysis, the dynamic evaluation of the vegetation restoration effect of the bedding rock slope is realized. The unmanned aerial vehicle remote sensing or ground measurement is used to obtain the slope vegetation information, including the green plant coverage area and the average root depth to calculate the repair evaluation value. The humidity sensors are used to collect the humidity data at multiple positions on the slope for a long time and stored in the form of a time-humidity data chain to construct a comprehensive humidity data set, avoiding the problem of insufficient accuracy caused by scattered data points in the traditional method. The Gaussian mixture distribution model is used to statistically analyze the humidity data, extract the most representative humidity data, and reduce the influence of single-point data fluctuations on the evaluation results. By comparing the current representative data chain with the historical humidity data set, the humidity change trend in the future period is predicted, making the evaluation method have foresight. When the predicted humidity data exceeds the safety threshold, the slope risk coefficient is further calculated. Combining with the slope stability theory and based on the influence of humidity on the shear strength of the slope, the repair evaluation value is dynamically adjusted to ensure the reliability of the evaluation results. Compared with the traditional static evaluation method, it can perceive the environmental changes in real time, early warn the risk of reduced slope stability, avoid the hidden dangers brought by overestimating the repair effect, and improve the safety and accuracy of slope management.

[0058] In another preferred manner based on the above embodiment, refer to Figure 3 As shown, this embodiment provides a vegetation restoration evaluation system for bedding rock slopes, which is used to apply the above-mentioned vegetation restoration evaluation method for bedding rock slopes, including: A collection unit, configured to collect the vegetation information of the slope to be evaluated, where the vegetation information includes the green plant area and the average root depth, and preliminarily determine the repair evaluation value according to the vegetation information.

[0059] A processing unit, configured to collect the humidity data at several locations on the slope to be evaluated within a preset period based on the humidity sensors, establish a time-humidity data chain for the data of each humidity sensor, construct a humidity data set from all the time-humidity data chains, and determine the representative humidity data at each moment of the slope to be evaluated based on the Gaussian mixture distribution and construct a representative data chain.

[0060] A judgment unit, configured to compare a representative data chain with a historical humidity data set, and determine predicted humidity data according to the comparison result. The historical humidity data set includes a number of historical representative data chains and a number of historical verified humidity data, and each historical representative data chain corresponds to a historical verified humidity data.

[0061] An adjustment unit, configured to compare the predicted humidity data with a humidity threshold, and determine whether to adjust the repair evaluation value according to the comparison result. When it is determined to adjust the repair evaluation value, determine a slope risk coefficient according to the predicted humidity data, and adjust the repair evaluation value according to the slope risk coefficient.

[0062] It can be understood that through vegetation information collection, humidity dynamic monitoring, data chain modeling, and risk assessment adjustment analysis, the dynamic assessment of the vegetation restoration effect of bedding rock slopes is realized. UAV remote sensing or ground measurement is used to obtain slope vegetation information, including the green plant coverage area and the average root depth to calculate the repair evaluation value. Humidity sensors are used to collect humidity data at multiple positions on the slope for a long time and stored in the form of a time-humidity data chain to construct a comprehensive humidity data set, avoiding the problem of insufficient accuracy caused by scattered data points in traditional methods. The Gaussian mixture distribution model is used to statistically analyze the humidity data, extract the most representative humidity data, and reduce the influence of single-point data fluctuations on the evaluation results. By comparing the current representative data chain with the historical humidity data set, the humidity change trend in the future period is predicted, making the evaluation method forward-looking. When the predicted humidity data exceeds the safety threshold, the slope risk coefficient is further calculated, combined with the slope stability theory, and the repair evaluation value is dynamically adjusted based on the influence of humidity on the shear strength of the slope to ensure the reliability of the evaluation results. Compared with traditional static evaluation methods, it can perceive environmental changes in real time, early warn of the risk of reduced slope stability, avoid the hidden dangers brought by overestimating the repair effect, and improve the safety and accuracy of slope management.

[0063] Those skilled in the art should understand that the embodiments of the present application can be provided as a method, a system, or a computer program product. Therefore, the present application can adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0064] This application is described with reference to the flowcharts and / or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the present application. It should be understood that each flow and / or block in the flowchart and / or block diagram can be implemented by computer program instructions, and the combination of flows and / or blocks in the flowchart and / or block diagram can also be implemented. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate means for implementing the functions specified in one flow Figure 1 one flow or multiple flows and / or blocks Figure 1 or multiple blocks.

[0065] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer-readable memory generate a manufactured article including instruction means, and the instruction means implement the functions specified in one flow Figure 1 one flow or multiple flows and / or blocks Figure 1 or multiple blocks.

[0066] These computer program instructions can also be loaded onto a computer or other programmable data processing device, so that a series of operation steps are executed on the computer or other programmable device to generate a computer-implemented process. Thus, the instructions executed on the computer or other programmable device provide steps for implementing the functions specified in one flow Figure 1 one flow or multiple flows and / or blocks Figure 1 or multiple blocks.

[0067] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit them. Although the present invention has been described in detail with reference to the above embodiments, those of ordinary skill in the art should understand that: modifications or equivalent replacements can still be made to the specific embodiments of the present invention, and any modification or equivalent replacement that does not depart from the spirit and scope of the present invention should be covered by the protection scope of the claims of the present invention.

Claims

1. A method for evaluating vegetation restoration of bedding rock slopes, characterized in that: include: Collecting vegetation information of the slope to be assessed, the vegetation information including green plant area and average root depth, and preliminarily determining the restoration assessment value based on the vegetation information; Based on humidity sensors, humidity data of several locations within a preset period of time on the slope to be evaluated are collected, data of each humidity sensor is established as a time-humidity data chain, all the time-humidity data chains are constructed as a humidity data set, representative humidity data of the slope to be evaluated at each moment is determined based on Gaussian mixture distribution and a representative data chain is constructed; Comparing the representative data chain with a historical humidity data set, and determining predicted humidity data according to the comparison result, the historical humidity data set comprising a plurality of historical representative data chains and a plurality of historical verification humidity data, and each of the historical representative data chains corresponds to a piece of historical verification humidity data; The predicted humidity data is compared with the humidity threshold, and it is determined whether to adjust the repair assessment value based on the comparison result. When it is determined that the repair assessment value is to be adjusted, the slope risk coefficient is determined based on the predicted humidity data, and the repair assessment value is adjusted based on the slope risk coefficient.

2. The method for evaluating vegetation restoration of bedding rock slopes according to claim 1 is characterized in that: The preliminary determination of the restoration assessment value according to the vegetation information includes: ； Among them, S represents the restoration assessment value, As represents the green plant area, Az represents the total slope area, ds represents the average root depth, dz represents the critical rooting depth, τs represents the measured root shear strength, and τz represents the reference root shear strength.

3. The method for evaluating vegetation restoration of bedding rock slopes according to claim 1, characterized in that: When determining the representative humidity data of the slope to be evaluated at each moment based on the Gaussian mixture distribution and constructing a representative data chain, it includes: The kernel density function expression is: Where n represents the total number of humidity data collected at the same time, h represents the smoothing bandwidth, represents the i-th humidity data collected at the same time, and x represents the value of a certain frequency point to be estimated; The humidity data corresponding to the highest frequency of kernel density is taken as the representative data at that moment; The representative data at each moment are combined to form the representative data chain.

4. The method for evaluating vegetation restoration of bedding rock slopes according to claim 3 is characterized in that: Comparing the representative data chain with the historical humidity data set and determining the predicted humidity data according to the comparison result includes: When there is data in the historical humidity data set whose similarity with the representative data chain is greater than the similarity threshold, the historical verified humidity number of the historical representative data chain corresponding to the maximum similarity is used as the predicted humidity data; When the similarities between all historical representative data chains in the historical humidity data set and the representative data chain are less than or equal to the similarity threshold, an LSTM model is trained based on the historical humidity data set, and the predicted humidity data is determined according to the LSTM model.

5. The method for evaluating vegetation restoration of bedding rock slopes according to claim 4 is characterized in that: When the predicted humidity data is determined according to the LSTM model, it includes: Each humidity data in the representative data chain is divided into windows according to the time series as input data, the output result of the LSTM model is obtained, and the output result is used as the predicted humidity data.

6. The method for evaluating vegetation restoration of bedding rock slopes according to claim 5 is characterized in that: When judging whether to adjust the restoration evaluation value according to the comparison result, it includes: The predicted humidity data is compared with a humidity threshold. When the predicted humidity data is greater than the humidity threshold, it is determined that the repair evaluation value is adjusted; when the predicted humidity data is less than or equal to the humidity threshold, it is determined that the repair evaluation value is not adjusted.

7. The method for evaluating vegetation restoration of bedding rock slopes according to claim 6 is characterized in that: When determining the slope risk factor based on the predicted humidity data, it includes: A humidity difference is obtained according to the predicted humidity data and the humidity threshold, the humidity difference is the difference between the predicted humidity data and the humidity threshold, the humidity difference is compared with a first preset difference and a second preset difference respectively, and the cohesion is determined according to the comparison result; the first preset difference is smaller than the second preset difference; When the humidity difference is less than or equal to a first preset difference, the cohesion is determined to be the first preset cohesion; when the humidity difference is greater than the first preset difference and less than the second preset difference, the cohesion is determined to be the second preset cohesion; when the humidity difference is greater than the second preset difference, the cohesion is determined to be the third preset cohesion; the first preset cohesion is greater than the second preset cohesion, the second preset cohesion is greater than the third preset cohesion, and the third preset cohesion is greater than or equal to 0.

8. The method for evaluating vegetation restoration of bedding rock slopes according to claim 7 is characterized in that: When determining the slope risk coefficient according to the predicted humidity data, it also includes: Among them, F is the slope risk coefficient, n is the cohesion, the height of the slope is H, the slope angle is α, the inclination angle of the structural surface is θ, γ is the weight of the rock mass, and φ is the internal friction angle.

9. The method for evaluating vegetation restoration of bedding rock slopes according to claim 8, characterized in that: When the restoration assessment value is adjusted according to the slope risk coefficient, it includes: The slope risk coefficient is compared with a minimum slope risk coefficient, the minimum slope risk coefficient being 1, and an adjustment coefficient is determined according to the comparison result to adjust the restoration assessment value; The adjustment coefficient is proportional to the slope risk coefficient, and the adjustment coefficient has a value range of (0, 1). The adjusted restoration assessment value is the product of the restoration assessment value and the adjustment coefficient.

10. A system for evaluating vegetation restoration of a bedding rock slope, used for applying the method for evaluating vegetation restoration of a bedding rock slope as claimed in any one of claims 1 to 9, characterized in that: include: A collection unit is configured to collect vegetation information of the slope to be evaluated, wherein the vegetation information includes green plant area and average root depth, and preliminarily determine a restoration evaluation value according to the vegetation information; A processing unit is configured to collect humidity data at a plurality of locations within a preset period of time of the slope to be evaluated based on a humidity sensor, establish the data of each humidity sensor as a time-humidity data chain, construct all the time-humidity data chains as a humidity data set, determine representative humidity data of the slope to be evaluated at each moment based on a Gaussian mixture distribution and construct a representative data chain; A judgment unit is configured to compare the representative data chain with a historical humidity data set, and determine predicted humidity data according to the comparison result, wherein the historical humidity data set includes a plurality of historical representative data chains and a plurality of historical verification humidity data, and each of the historical representative data chains corresponds to a piece of historical verification humidity data; The adjustment unit is configured to compare the predicted humidity data with the humidity threshold, determine whether to adjust the repair assessment value based on the comparison result, and when it is determined that the repair assessment value is to be adjusted, determine the slope risk coefficient based on the predicted humidity data, and adjust the repair assessment value based on the slope risk coefficient.

Images