Foundation pit soil body reinforcing method and equipment based on electrical parameter monitoring and medium

Abstract

The invention provides a foundation pit soil body reinforcement method based on electrical parameter monitoring, and relates to the technical field of geotechnical engineering monitoring and automatic reinforcement, and the method comprises the following steps: burying an electrode array in a foundation pit soil body to collect electrical parameters such as resistivity and capacitance value, carrying out normalization, anomaly detection and space-time alignment processing, and determining the electrical parameters in the foundation pit soil body; generating soil state data through an adaptive weighted fusion algorithm; constructing a stability model based on the fused data to calculate a stability index, smoothing fluctuation by adopting moving average filtering, predicting future change through a GRU model, and triggering early warning when the future change is lower than a safety baseline; dividing risk levels according to the stability index and the deformation factor, and generating a reinforcement instruction when the risk is high; the comprehensive reinforcement priority is calculated by combining the emergency degree and the regional importance determined by the entropy weight method, and the resource demand is evaluated by applying a multi-objective optimization algorithm; and an ant colony optimization algorithm is used for dynamically optimizing the reinforcement scheme, and automatic grouting reinforcement is achieved. The construction safety and the resource utilization efficiency can be improved.

Description

Technical Field

[0001] The present invention relates to the technical field of geotechnical engineering monitoring and automatic reinforcement, and more specifically, the present invention relates to a method, device, and medium for foundation pit soil reinforcement based on electrical parameter monitoring. Background Art

[0002] In foundation pit engineering, soil reinforcement is a key link to ensure construction safety and quality. With the acceleration of the urbanization process, there are more and more deep foundation pit projects, and their construction environment is complex, and the problem of soil stability is particularly prominent. Traditional soil reinforcement methods mainly rely on empirical judgment and simple monitoring means, such as evaluating the soil state by manual observation, regular sampling and testing, etc. These methods have many limitations, such as incomplete monitoring data, poor real-time performance, inability to accurately predict changes in soil stability, etc., resulting in difficulty in timely discovering potential soil instability risks in actual construction, thereby affecting the construction progress and safety.

[0003] In recent years, with the development of sensor technology and data processing technology, soil reinforcement methods based on electrical parameter monitoring have gradually attracted attention. Electrical parameters such as resistivity, capacitance value, and dielectric constant are closely related to physical properties such as soil moisture content, particle structure, and compaction degree, and can reflect the microscopic structure changes and stability state of the soil. By burying an electrode array in the foundation pit soil, electrical parameter data of the soil can be collected in real time, providing a more accurate basis for soil reinforcement. However, there are still deficiencies in the processing and analysis of electrical parameter data in the prior art. On the one hand, the collected electrical parameter data is often affected by noise interference, and there is a problem of inconsistent data ranges between different parameters, resulting in difficult data fusion and inability to accurately reflect the true state of the soil. On the other hand, the prior art lacks an effective prediction model for changes in soil stability, unable to early warn of soil instability risks, and difficult to achieve a dynamically optimized reinforcement plan.

[0004] In the process of implementing the embodiments of the present invention, the inventors found that there are at least the following problems or defects in the prior art: The existing soil reinforcement methods do not process electrical parameter data finely enough, unable to effectively remove noise and accurately fuse multi-source data; lack the ability to dynamically predict changes in soil stability, difficult to early warn and dynamically adjust the reinforcement strategy; the optimization of the reinforcement plan lacks scientific resource allocation and priority ranking methods, resulting in low reinforcement efficiency and serious resource waste. Summary of the Invention

[0005] The present invention provides a method, device, and medium for foundation pit soil reinforcement based on electrical parameter monitoring.

[0006] In the first aspect of the present invention, a method for foundation pit soil reinforcement based on electrical parameter monitoring is provided, including: An electrode array is buried in the foundation pit soil to collect electrical parameter data, and the data is aligned in time and space. The aligned data is used to generate soil state data through an adaptive weighted fusion algorithm, and data re-collection and calibration are triggered when the fluctuation exceeds the threshold. A soil stability model is constructed based on the fused data, the stability index is calculated to evaluate the soil state, short-term fluctuations are smoothed using a sliding average filter, future stability changes are predicted using a GRU model, and an early warning is triggered when the soil is below the safety baseline. The risk level is divided according to the stability index and deformation factor, and reinforcement instructions are generated when the risk is high; The reinforcement priority is set based on stability assessment and deformation prediction, the urgency is calculated by combining the stability index and deformation risk, and the regional importance is determined by the entropy weight method. The multi-objective optimization algorithm is applied to evaluate the reinforcement resource requirements and dynamically monitor the material consumption. Use ant colony optimization algorithm to dynamically optimize the reinforcement scheme.

[0007] Furthermore, an electrode array is buried in the foundation pit soil to collect electrical parameter data, and the data is subjected to spatiotemporal alignment processing. The aligned data is used to generate soil state data through an adaptive weighted fusion algorithm, and data re-collection and calibration are triggered when the fluctuation exceeds the threshold. The specific steps are as follows: The data collected by the electrode array include resistivity parameters, capacitance parameters, and dielectric constant parameters; Normalize the collected data to make the data range of different electrical parameters consistent; The LOF local outlier factor algorithm is used to detect anomalies on each normalized data to remove noise data, and the collected data is aligned in time and space to align data at different spatial locations to the same time series; After completing the time-space alignment, the collected data are fused using an adaptive weighted fusion algorithm based on dynamic weight allocation, where the weight allocation is dynamically adjusted according to the historical correlation between each electrical parameter and soil stability; The fused data is used as the soil state input at the current time point, and a feedback mechanism is set. When the difference between the fused value and the value at the previous time point exceeds the set threshold, the electrode array data acquisition is restarted and the measuring equipment is calibrated.

[0008] Furthermore, a soil stability model is constructed based on the fused data, the stability index is calculated to evaluate the soil state, and a moving average filter is used to smooth short-term fluctuations. The specific steps include: The fused data is used as the input data for stability assessment. A soil stability model is established for the fused data. The stability index of the soil is calculated by multivariate regression analysis and covariance matrix generation. The stability state of the soil is determined based on the stability index. Use sliding average filtering to dynamically smooth the soil stability index to deal with the impact of short-term fluctuations; For the region where the steady state is at the critical value, the deviation degree between the eigenvalue of the fusion data and the safety threshold is extracted to determine the deformation factor, which is used to analyze the structural risk.

[0009] Furthermore, the GRU model is used to predict the future stability change, and an early warning is triggered when it is lower than the safety baseline. The specific steps are as follows: Build a deformation prediction model based on historical data, and use the GRU model to predict the stability index at future moments; The input of the deformation prediction model is the current and historical stability indices and the deformation factor; If the predicted stability index continues to be lower than the safety baseline threshold, an early warning signal is triggered to indicate the risk of soil instability.

[0010] Furthermore, the risk level is divided according to the stability index and the deformation factor, and a reinforcement instruction is generated when it is at a high risk. The steps include: Determine the risk level of each monitoring area according to the decline rate of the stability index and the cumulative amount of the deformation factor. The risk levels include low risk level, warning risk level and high risk level; When the risk level is the low risk level, no reinforcement instruction is generated; When the risk level is the warning risk level, mark the monitoring area as a warning area and formulate a monitoring strengthening plan; When the risk level is the high risk level, generate a reinforcement instruction signal to trigger the automatic grouting system for active reinforcement; Among them, the conditions for the low risk level are: The decline rate is lower than the first decline rate threshold, and the cumulative amount of the deformation factor is lower than the first deformation factor cumulative amount threshold; The conditions for the warning risk level are: The decline rate reaches or exceeds the first decline rate threshold but is lower than the second decline rate threshold, and the cumulative amount of the deformation factor is lower than the first deformation factor cumulative amount threshold; or, the cumulative amount of the deformation factor reaches or exceeds the first deformation factor cumulative amount threshold but is lower than the second deformation factor cumulative amount threshold, and the decline rate is lower than the first decline rate threshold; The conditions for the high risk level are: The decline rate reaches or exceeds the second decline rate threshold; or, the cumulative amount of the deformation factor reaches or exceeds the second deformation factor cumulative amount threshold; or, the decline rate reaches or exceeds the first decline rate threshold and the cumulative amount of the deformation factor reaches or exceeds the first deformation factor cumulative amount threshold.

[0011] Furthermore, based on the stability assessment and deformation prediction, set the reinforcement priority, calculate the urgency by combining the stability index and deformation risk, and determine the regional importance through the entropy weight method, including the following steps: According to the stability assessment and deformation prediction results, quantify the urgency of the reinforcement requirements for each monitoring area. The urgency index is used to represent the intensity of the reinforcement requirements in the area at the current moment; Determine the relative importance of each monitoring area through the entropy weight method, construct a regional importance judgment matrix, and assign values according to the spatial relationship between the area and the support structure; Calculate the weight distribution of the judgment matrix by the coefficient of variation method, standardize the calculation results, and obtain the weight of each area as the regional importance index; After the calculation, determine the key reinforcement areas according to the obtained regional importance index values; After obtaining the task urgency and regional importance, calculate the comprehensive reinforcement priority index, and the expression formula is:

[0012] Among them, is the comprehensive reinforcement priority index of area i; is the adjustment parameter, satisfying ; is the urgency index of area i, calculated by the difference between the current stability index and the baseline threshold; represents the cumulative amount of deformation factors in area i at historical time t; is the attenuation coefficient, used to adjust the influence weight of historical deformation factors; is the importance index of area i, the weight value standardized by the entropy weight method; is the base of the exponential function; Sort the reinforcement tasks of each area according to the comprehensive reinforcement priority index, generate a reinforcement task queue, and the task sorting follows the principle of descending priority. The area at the head of the queue is reinforced first.

[0013] Furthermore, apply the multi-objective optimization algorithm to evaluate the reinforcement resource requirements and dynamically monitor the material consumption, including the following steps: Based on the priority and reinforcement volume of each reinforcement task, evaluate the required amount of grouting material; According to the reinforcement task type and soil properties, determine the curing agent ratio plan required for each task, and combine the material inventory limit and supply capacity to calculate the material scheduling plan that meets the requirements of all tasks; For each reinforcement task, analyze the equipment requirements, including the grouting machine model and drilling equipment, and calculate the equipment occupation time; Based on the task priority index and material requirements, equipment requirements, and manpower requirements, and applying the ant colony optimization algorithm for dynamic resource scheduling to determine the resource scheduling cost.

[0014] Furthermore, the ant colony optimization algorithm is used to dynamically optimize the reinforcement plan, including the following steps: Construct a multi-objective optimization model. Based on resource constraints and task priorities, construct a multi-objective optimization model. The objective function of the model includes material consumption cost, reinforcement completion time, and resource utilization rate. The resources to be scheduled include grouting materials, construction equipment, and monitoring personnel. Select the ant colony optimization algorithm to optimize the reinforcement execution path.

[0015] In the second aspect of the present invention, an electronic device is provided. The electronic device includes: at least one processor, a memory, and an input-output unit; wherein, the memory is used to store a computer program, and the processor is used to call the computer program stored in the memory to execute the method described in any one of the first aspect.

[0016] In the third aspect of the present invention, a computer-readable storage medium is provided, which includes instructions that, when running on a computer, cause the computer to execute the method described in any one of the first aspect.

[0017] According to the above embodiments of the present invention, it has at least the following beneficial effects: By burying an electrode array in the foundation pit soil body to collect electrical parameter data and using an adaptive weighted fusion algorithm to process the data, the present invention can effectively solve the problems of difficult data fusion and noise interference in the prior art. By performing normalization processing, anomaly detection, and spatio-temporal alignment on the collected data, the accuracy and consistency of the data can be ensured, providing reliable data support for the accurate assessment of the soil body state. At the same time, the fusion algorithm based on dynamic weight allocation can dynamically adjust the weights according to the historical correlation between each electrical parameter and soil body stability, further improving the accuracy of data fusion. In addition, the present invention uses moving average filtering and GRU model to dynamically evaluate and predict the soil body stability, which can smooth short-term fluctuations and early warning of the risk of soil body instability, providing sufficient time for construction personnel to take reinforcement measures, thereby reducing the safety risk of foundation pit engineering.

[0018] The present invention can also dynamically optimize the reinforcement plan through the entropy weight method and the multi-objective optimization algorithm, reasonably allocate reinforcement resources, and determine the reinforcement priority. By calculating the urgency index and the regional importance index, and combining the comprehensive reinforcement priority index to sort the reinforcement tasks, it can ensure that the key areas are reinforced first, improving the reinforcement efficiency. At the same time, the dynamic resource scheduling based on the ant colony optimization algorithm can optimize the allocation of grouting materials, construction equipment, and monitoring personnel, reduce the resource scheduling cost, and improve the resource utilization rate. This dynamically optimized reinforcement plan can not only improve the scientificity and accuracy of the foundation pit soil reinforcement, but also effectively save the construction cost, improve the construction efficiency, and provide guarantee for the safe construction of the foundation pit project. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] By referring to the following detailed description with reference to the accompanying drawings, the above and other objects, features, and advantages of the exemplary embodiments of the present invention will become readily understood. In the drawings, several embodiments of the present invention are shown by way of example and not limitation, wherein: Figure 1 FIG. is a schematic flowchart of a foundation pit soil reinforcement method based on electrical parameter monitoring provided by an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0020] The principles and spirit of the present invention will be described below with reference to several exemplary embodiments. It should be understood that these embodiments are provided only to enable those skilled in the art to better understand and then implement the present invention, and do not limit the scope of the present invention in any way. On the contrary, these embodiments are provided to make the present invention more thorough and complete, and to be able to convey the scope of the present invention fully to those skilled in the art.

[0021] Those skilled in the art know that the embodiments of the present invention can be implemented as a system, device, equipment, method, or computer program product. Therefore, the present invention can be specifically implemented in the following forms: completely hardware, completely software (including firmware, resident software, microcode, etc.), or a combination of hardware and software.

[0022] It should be noted that any number of elements in the drawings is for illustration rather than limitation, and any naming is only for distinction and does not have any limiting meaning.

[0023] Refer to the following Figure 1 , Figure 1 FIG. is a schematic flowchart of a foundation pit soil reinforcement method based on electrical parameter monitoring provided by an embodiment of the present invention. As Figure 1 shown, a foundation pit soil reinforcement method based on electrical parameter monitoring includes: S1. Install an electrode array in the foundation pit soil mass to collect electrical parameter data, perform spatio-temporal alignment processing on the data, generate soil mass state data through an adaptive weighted fusion algorithm, and trigger re-collection and calibration of the data when the fluctuation exceeds the threshold; S2. Construct a soil mass stability model based on the fusion data, calculate the stability index to evaluate the soil mass state, use moving average filtering to smooth short-term fluctuations, predict future stability changes through a GRU model, trigger an early warning when it is below the safety baseline, divide the risk level according to the stability index and deformation factor, and generate a reinforcement instruction when it is at high risk; S3. Set the reinforcement priority based on stability assessment and deformation prediction, calculate the urgency by combining the stability index and deformation risk, determine the regional importance through the entropy weight method, evaluate the reinforcement resource requirements using a multi-objective optimization algorithm, and dynamically monitor the material consumption; S4. Dynamically optimize the reinforcement plan using the ant colony optimization algorithm.

[0024] It should be noted that during the foundation pit soil mass reinforcement process, it is first necessary to install an electrode array in the foundation pit soil mass to collect electrical parameter data. The electrode array is a sensor network used to measure the electrical properties of the soil mass, which can obtain parameters such as the resistivity, capacitance value, and dielectric constant of the soil mass in real time. These parameters are closely related to the physical properties of the soil mass such as water content, particle structure, and compaction degree, and can reflect the microscopic structure changes and stability state of the soil mass. The collected data needs to be subjected to spatio-temporal alignment processing, that is, aligning the data at different spatial positions to the same time series to ensure the consistency and comparability of the data. The adaptive weighted fusion algorithm is a data fusion method that dynamically adjusts weights, which can dynamically allocate weights according to the historical correlation between each electrical parameter and the soil mass stability, so as to generate more accurate soil mass state data. When the difference between the fusion value and the value at the previous time point exceeds the set threshold, a data re-collection and calibration mechanism will be triggered to ensure the accuracy of the measuring device and the reliability of the data. Here, the spatio-temporal alignment processing refers to uniformly processing the data collected at different spatial positions in time to ensure the consistency of the data in the time dimension. And the adaptive weighted fusion algorithm is an intelligent data processing method that can dynamically adjust the weights of each parameter according to the historical performance of the data to better reflect the actual state of the soil mass.

[0025] Specifically, the data collected by the electrode array include electrical parameters such as resistivity, capacitance value, and dielectric constant. Resistivity refers to the ability of the soil mass to resist the flow of electric current and is usually related to the water content and particle structure of the soil mass; the capacitance value reflects the energy storage ability of the soil mass in an electric field and is closely related to the pore structure and water content of the soil mass; the dielectric constant represents the degree of polarization of the soil mass in an electric field and is related to the composition and structure of the soil mass. After these parameters are collected by the electrode array, they first need to be normalized, that is, the data ranges of different parameters are unified into the same interval, such as [0, 1], for subsequent fusion processing. After normalization, the Local Outlier Factor (LOF) algorithm is used to detect anomalies in the data and eliminate noise data. The spatio-temporal alignment process aligns the electrical parameter data at different spatial positions onto the same time series to ensure the time consistency of the data. The adaptive weighted fusion algorithm dynamically adjusts the weights according to the historical correlation between each electrical parameter and the soil stability. The specific way of weight allocation can be obtained through statistical analysis of historical data. For example, if the resistivity has a high correlation with the soil stability, a larger weight is assigned. The threshold is a parameter used to judge whether the data fluctuation is abnormal. When the difference between the fusion value and the value at the previous time point exceeds this threshold, data re-acquisition and calibration are triggered. Here, the threshold refers to a preset value used to judge whether the data change is within the normal range. If it exceeds this threshold, it is considered that the data may be abnormal and needs to be re-acquired and calibrated.

[0026] Preferably, when constructing the soil stability model, methods such as multiple regression analysis and covariance matrix generation can be used to calculate the soil stability index. Multiple regression analysis is a statistical method that establishes a mathematical model by analyzing the relationship between multiple independent variables, electrical parameters, and the dependent variable, soil stability. The covariance matrix is used to describe the correlation between different electrical parameters. By calculating the eigenvalues and eigenvectors of the covariance matrix, the influence weights of each parameter on the soil stability can be obtained. The stability index is a quantitative indicator to measure the stable state of the soil mass, and the higher its value, the more stable the soil mass. When performing dynamic smoothing processing on the stability index, the moving average filtering method can be used to smooth short-term fluctuations by calculating the average value within a certain time window. For the region where the stable state is at the critical value, the deformation factor can be determined by extracting the deviation degree of the eigenvalue of the fusion data from the safety threshold. The deformation factor is used to analyze the structural risk, and the larger its value, the higher the deformation risk of the soil mass. The moving average filtering is a commonly used data smoothing method that reduces short-term fluctuations in the data by calculating the average value over a period of time, so as to more clearly observe the long-term trend of the data. The deformation factor is calculated based on the deviation degree of the fusion data from the safety threshold and is a quantitative indicator used to evaluate the deformation risk of the soil mass, which can help judge whether the soil mass is approaching the unstable state.

[0027] In some embodiments, an electrode array is buried in the foundation pit soil to collect electrical parameter data, and the data is subjected to spatiotemporal alignment processing. The aligned data is used to generate soil state data through an adaptive weighted fusion algorithm, and data re-collection and calibration are triggered when the fluctuation exceeds a threshold. The specific steps are as follows: The data collected by the electrode array include resistivity parameters, capacitance parameters, and dielectric constant parameters; Normalize the collected data to make the data range of different electrical parameters consistent; The LOF local outlier factor algorithm is used to detect anomalies on each normalized data to remove noise data, and the collected data is aligned in time and space to align data at different spatial locations to the same time series; After completing the time-space alignment, the collected data are fused using an adaptive weighted fusion algorithm based on dynamic weight allocation, where the weight allocation is dynamically adjusted according to the historical correlation between each electrical parameter and soil stability; The fused data is used as the soil state input at the current time point, and a feedback mechanism is set. When the difference between the fused value and the value at the previous time point exceeds the set threshold, the electrode array data acquisition is restarted and the measuring equipment is calibrated.

[0028] It should be noted that in the process of soil reinforcement of foundation pit, collecting electrical parameter data by burying electrode arrays in the soil of foundation pit and performing spatiotemporal alignment processing on the data is the basis for accurate monitoring of soil status. The data collected by the electrode array include parameters such as resistivity, capacitance and dielectric constant, which can reflect the physical properties and stability state of the soil. Normalization processing is to unify the data range of different electrical parameters to the same interval for subsequent fusion processing. The local outlier factor LOF algorithm is used to detect outliers in the data, eliminate noise data, and ensure the reliability of the data. Spatiotemporal alignment processing is to align the data collected at different spatial locations to the same time series to ensure the temporal consistency of the data. The adaptive weighted fusion algorithm dynamically adjusts the weights according to the historical correlation between each electrical parameter and soil stability to generate more accurate soil status data. When the difference between the fusion value and the value at the previous time point exceeds the set threshold, the data re-collection and calibration mechanism is triggered to ensure the accuracy of the measuring equipment and the reliability of the data. These steps together constitute a complete data processing process, providing data support for subsequent soil stability assessment and reinforcement plans.

[0029] Specifically, the data collected by the electrode array includes electrical parameters such as resistivity, capacitance value, and dielectric constant. Resistivity refers to the ability of the soil to resist the flow of electric current, which is usually related to the water content and particle structure of the soil; the capacitance value reflects the energy storage ability of the soil in an electric field and is closely related to the pore structure and water content of the soil; the dielectric constant represents the degree of polarization of the soil in an electric field and is related to the composition and structure of the soil. Normalization is to unify the data ranges of different parameters into the same interval, such as [0, 1], for subsequent fusion processing. The Local Outlier Factor (LOF) algorithm is a density-based anomaly detection algorithm that determines whether a data point is an outlier by calculating its local density. Spatiotemporal alignment processing is to align the electrical parameter data at different spatial positions onto the same time series to ensure the time consistency of the data. The adaptive weighted fusion algorithm dynamically adjusts the weights according to the historical correlation between each electrical parameter and soil stability, and the specific method of weight allocation can be obtained based on historical data statistics. The threshold is a parameter used to determine whether the data fluctuation is abnormal. When the difference between the fusion value and the value at the previous time point exceeds this threshold, data re-acquisition and calibration are triggered. For example, the threshold can be set to 0.1, indicating that when the data change exceeds 10%, it is considered that the data may be abnormal and needs to be re-acquired and calibrated.

[0030] Preferably, the normalization process can be achieved through the following steps: First, calculate the maximum and minimum values of each electrical parameter, and then divide the value of each data point minus the minimum value by the difference between the maximum and minimum values, so as to unify the data range into the [0, 1] interval. The implementation process of the Local Outlier Factor (LOF) algorithm includes: calculating the neighborhood density of each data point, comparing its density difference with other points in the neighborhood, and if the difference is large, then this point is considered an outlier. Spatiotemporal alignment processing can be achieved through interpolation methods, such as linear interpolation or spline interpolation, to align the data at different spatial positions onto the same time series. The weight allocation of the adaptive weighted fusion algorithm can be obtained based on the correlation analysis of historical data. For example, by calculating the correlation coefficient between each electrical parameter and the soil stability index, the correlation coefficient is used as the basis for weight allocation. When the difference between the fusion value and the value at the previous time point exceeds the set threshold, the specific process of triggering data re-acquisition and calibration is: First, pause the current data acquisition, restart the electrode array for data acquisition, and calibrate the measurement device to ensure the accuracy and reliability of the data.

[0031] In some embodiments, a soil stability model is constructed based on the fusion data, the stability index is calculated to evaluate the soil state, and a moving average filter is used to smooth short-term fluctuations. The specific steps include: The fusion data is used as the input data for stability evaluation. A soil stability model is established for the fusion data, and the stability index of the soil is calculated through multiple regression analysis and the generation of covariance matrices. The stable state of the soil is determined according to the stability index; Use moving average filtering to dynamically smooth the short-term fluctuation effects on the soil stability index; For the regions where the stable state is at the critical value, extract the deviation degree between the eigenvalue of the fused data and the safety threshold to determine the deformation factor, which is used to analyze the structural risk.

[0032] It should be noted that in the process of foundation pit soil reinforcement, constructing a soil stability model based on fused data is a key step to achieve accurate assessment of the soil state. Fused data refers to the electrical parameter data after normalization processing, anomaly detection, and spatio-temporal alignment. These data can more accurately reflect the actual state of the soil. By calculating the soil stability index through multiple regression analysis and covariance matrix generation methods, the stable state of the soil can be quantified. The stability index is a comprehensive indicator used to evaluate the stability level of the soil under current conditions. Moving average filtering is a commonly used data smoothing technique. By dynamically smoothing the stability index, the influence of short-term fluctuations can be effectively reduced, making the stability assessment more stable and reliable. For the regions where the stable state is at the critical value, extract the deviation degree between the eigenvalue of the fused data and the safety threshold to determine the deformation factor, which is used to analyze the structural risk and help identify potential unstable regions.

[0033] Specifically, fused data refers to the electrical parameter data after a series of preprocessing, including normalization processing, anomaly detection, and spatio-temporal alignment. Normalization processing unifies the data ranges of different electrical parameters to the interval [0, 1] for subsequent fusion processing. Anomaly detection eliminates noise data through the Local Outlier Factor (LOF) algorithm to ensure the reliability of the data. Spatio-temporal alignment aligns the data at different spatial positions to the same time series to ensure the time consistency of the data. Multiple regression analysis is a statistical method that establishes a mathematical model by analyzing the relationships between multiple independent variables, electrical parameters, and the dependent variable, soil stability. The covariance matrix is used to describe the correlations between different electrical parameters. By calculating the eigenvalues and eigenvectors of the covariance matrix, the influence weights of each parameter on soil stability can be obtained. The stability index is calculated through multiple regression analysis and covariance matrix generation methods and is used to quantify the stable state of the soil. Moving average filtering is a data smoothing technique that reduces the influence of short-term fluctuations by dynamically smoothing the stability index. The deformation factor is determined by extracting the deviation degree between the eigenvalue of the fused data and the safety threshold and is used to analyze the structural risk. The larger the value, the higher the soil deformation risk.

[0034] Preferably, the specific steps for constructing the soil stability model are as follows: First, use the preprocessed fusion data as input, and establish a relationship model between electrical parameters and soil stability through multiple regression analysis. In multiple regression analysis, the input parameters include electrical parameters such as resistivity, capacitance value, and dielectric constant, and the output is the stability index. Then, calculate the covariance matrix to describe the correlation between different electrical parameters. Through the analysis of the eigenvalues and eigenvectors of the covariance matrix, determine the influence weights of each electrical parameter on soil stability. The calculation formula of the stability index can be expressed as: Stability index = w1×resistivity + w2×capacitance value + w3×dielectric constant, where w1, w2, and w3 are the weight coefficients obtained from the covariance matrix analysis, w1, w2, and w3 ∈ [0, 1], and the sum of w1, w2, and w3 is 1. The specific implementation process of moving average filtering is: select a time window, for example, 5 time points, calculate the average value of the stability index within this time window, and then move the time window point by point forward and repeat the calculation of the average value to achieve dynamic smoothing processing of the stability index. For the area where the stable state is at the critical value, the calculation method of the deformation factor is: extract the eigenvalue of the fusion data and calculate its deviation from the safety threshold. The greater the deviation, the greater the deformation factor, indicating a higher deformation risk in this area.

[0035] In some embodiments, the future stability change is predicted through the GRU model, and an early warning is triggered when it is lower than the safety baseline. The specific steps are as follows: Construct a deformation prediction model based on historical data, and apply the GRU model to predict the stability index at future moments; The input of the deformation prediction model is the current and historical stability indices and deformation factors; If the predicted stability index continuously falls below the safety baseline threshold, an early warning signal is triggered to indicate the risk of soil instability.

[0036] It should be noted that in the process of foundation pit soil reinforcement, predicting the future stability change through the GRU model is a key link to achieve early warning and dynamic reinforcement. The GRU, Gated Recurrent Unit, model is an improved recurrent neural network RNN structure, specifically used to process time series data and capable of capturing long-term dependencies in the data. In the present invention, the input of the GRU model is the current and historical stability indices and deformation factors, and these parameters can reflect the current state and historical change trend of the soil. By predicting the stability index at future moments, it is possible to judge in advance whether there is a risk of soil instability. When the predicted stability index continuously falls below the safety baseline threshold, an early warning signal is triggered to indicate the risk of soil instability. This early warning mechanism can provide sufficient time for construction personnel to take measures to avoid potential safety accidents.

[0037] Specifically, the GRU model is a recurrent neural network based on a gating mechanism, which can effectively solve the problem of gradient vanishing or gradient explosion in traditional RNNs when dealing with long sequence data. The GRU model controls the flow of information through an update gate and a reset gate, thereby better capturing long-term dependencies in time series data. In the present invention, the inputs of the GRU model include current and historical stability indices and deformation factors. The stability index is calculated by means of multiple regression analysis and covariance matrix generation, and is used to quantify the stable state of the soil mass; the deformation factor is determined by extracting the degree of deviation of the eigenvalue of the fusion data from the safety threshold, and is used to analyze structural risks. The safety baseline threshold is a preset stability index value, which is used to judge whether the soil mass is in a safe state. When the predicted stability index is lower than this threshold, it indicates that there may be a risk of instability in the soil mass, and measures need to be taken in a timely manner. For example, the safety baseline threshold can be set to a specific value, such as 0.8, according to historical data and experience, indicating that when the stability index is lower than 0.8, a warning is triggered.

[0038] Preferably, the specific steps for constructing the GRU model are as follows: First, collect and organize the historical stability index and deformation factor data of the foundation pit soil mass as the input of the model. The input data needs to be standardized to ensure the training effect of the model. Then, select an appropriate GRU network structure, including determining the number of units and the number of layers in the hidden layer. For example, the number of hidden layer units can be set to 50 and the number of layers to 2 to ensure that the model can fully capture the complex relationships in the data. Next, use the historical data to train the GRU model, and adjust the model parameters through an optimization algorithm such as Adam so that the model can accurately predict the future stability index. After the model training is completed, use the trained GRU model to predict the stability index and deformation factor collected in real time. During the prediction process, the model will output the stability index at a future time according to the input current and historical data. Finally, compare the predicted stability index with the safety baseline threshold. If the predicted value is continuously lower than the threshold, trigger a warning signal to remind the construction personnel to take reinforcement measures. The warning signal can be sent in various ways, such as by text message, alarm sound or visual interface display, to ensure that the construction personnel can receive the warning information in a timely manner and take corresponding measures.

[0039] In some embodiments, risk levels are divided according to the stability index and deformation factor, and reinforcement instructions are generated when the risk is high, including the following steps: Determine the risk level of each monitoring area according to the decline rate of the stability index and the cumulative amount of the deformation factor. The risk levels include low risk level, warning risk level and high risk level; When the risk level is the low risk level, no reinforcement instruction is generated; When the risk level is the warning risk level, mark the monitored area as a warning area and formulate a monitoring enhancement plan; When the risk level is the high risk level, generate a reinforcement instruction signal to trigger the automatic grouting system for active reinforcement; Among them, the conditions for the low risk level are: The rate of decline is lower than the first rate-of-decline threshold, and the cumulative amount of the deformation factor is lower than the first cumulative deformation factor threshold; The conditions for the warning risk level are: The rate of decline reaches or exceeds the first rate-of-decline threshold but is lower than the second rate-of-decline threshold, and the cumulative amount of the deformation factor is lower than the first cumulative deformation factor threshold; or, the cumulative amount of the deformation factor reaches or exceeds the first cumulative deformation factor threshold but is lower than the second cumulative deformation factor threshold, and the rate of decline is lower than the first rate-of-decline threshold; The conditions for the high risk level are: The rate of decline reaches or exceeds the second rate-of-decline threshold; or, the cumulative amount of the deformation factor reaches or exceeds the second cumulative deformation factor threshold; or, the rate of decline reaches or exceeds the first rate-of-decline threshold and the cumulative amount of the deformation factor reaches or exceeds the first cumulative deformation factor threshold.

[0040] It should be noted that during the foundation pit soil reinforcement process, the risk level is divided based on the rate of decline of the stability index and the cumulative amount of the deformation factor. The rate of decline and the cumulative amount of the deformation factor can comprehensively reflect the current state and potential risks of the soil mass. The risk levels include low risk level, warning risk level, and high risk level: when the risk level is the low risk level, it means the soil mass state is normal and no immediate reinforcement is required; when the risk level is the warning risk level, mark the monitored area as a warning area and formulate a monitoring enhancement plan; when the risk level is the high risk level, trigger a reinforcement instruction signal to start the automatic grouting system for active reinforcement. Through the hierarchical early warning mechanism, the timeliness of reinforcement and the reasonable utilization of resources can be balanced to ensure the safety and economy of the foundation pit project.

[0041] Specifically, the stability index is calculated by means of multiple regression analysis and covariance matrix generation, and is used to quantify the stable state of the soil mass. The higher its value, the more stable the soil mass. The rate of decline is the change rate of the stability index over time, and is determined by calculating the ratio of the difference in the stability index between adjacent time points to the time interval; the deformation factor is determined by extracting the deviation degree of the eigenvalue of the fusion data from the safety threshold and is used to analyze the structural risk; the cumulative amount of the deformation factor is the cumulative value of the deformation factor over a period of time, reflecting the cumulative degree of the soil mass deformation risk.

[0042] Specifically, the basis for determining the risk level is as follows: when the descent rate does not reach the first descent rate threshold and the cumulative amount of the deformation factor does not reach the first cumulative deformation factor threshold, it is determined as a low risk level, and no reinforcement instruction is generated at this time; if the descent rate reaches or exceeds the first descent rate threshold but does not reach the second descent rate threshold, and the cumulative amount of the deformation factor does not reach the first cumulative deformation factor threshold, or the cumulative amount of the deformation factor reaches or exceeds the first cumulative deformation factor threshold but does not reach the second cumulative deformation factor threshold and the descent rate does not reach the first descent rate threshold, it is determined as a warning risk level, the monitored area is marked as a warning area and a monitoring intensification plan is formulated; if the descent rate reaches or exceeds the second descent rate threshold, or the cumulative amount of the deformation factor reaches or exceeds the second cumulative deformation factor threshold, or the descent rate reaches or exceeds the first descent rate threshold and the cumulative amount of the deformation factor simultaneously reaches or exceeds the first cumulative deformation factor threshold, it is determined as a high risk level, a reinforcement instruction signal is generated and the automatic grouting system is triggered for active reinforcement.

[0043] Preferably, the specific steps for risk level classification are as follows: calculate the descent rate of the stability index and the cumulative amount of the deformation factor in real time, where the formula for the descent rate is (the current stability index minus the stability index at the previous time point) divided by the time interval, and the cumulative amount of the deformation factor is the cumulative value of the deformation factor at historical moments; determine the risk level according to the calculation results. If both the descent rate and the cumulative amount of the deformation factor do not reach the first threshold, it is determined as a low risk. If only a single parameter (the descent rate or the cumulative amount of the deformation factor) is between the first threshold and the second threshold and the other parameter does not reach the first threshold, it is determined as a warning risk. If any parameter exceeds the second threshold or both parameters exceed the first threshold simultaneously, it is determined as a high risk. The reinforcement instruction signal is automatically sent through the control system to ensure that reinforcement measures are taken in a timely manner in high-risk areas and the safety of the foundation pit project is guaranteed.

[0044] In some embodiments, based on the stability assessment and deformation prediction, a reinforcement priority is set, the urgency is calculated by combining the stability index and the deformation risk, and the regional importance is determined by the entropy weight method, including the following steps: According to the stability assessment and deformation prediction results, quantify the urgency of the reinforcement requirements for each monitored area, and the urgency index is used to represent the intensity of the reinforcement requirements for the area at the current moment; Determine the relative importance of each monitored area by the entropy weight method, construct a regional importance judgment matrix, and assign values according to the spatial relationship between the area and the support structure; Calculate the weight distribution of the judgment matrix by the coefficient of variation method, standardize the calculation results, and obtain the weight of each area as the regional importance index; After the calculation is completed, determine the key reinforcement areas according to the obtained regional importance index values; After obtaining the task urgency level and regional importance, calculate the comprehensive reinforcement priority index, and the expression formula is:

[0045] Among them, is the comprehensive reinforcement priority index of area i; is the adjustment parameter, satisfying ; is the urgency index of area i, calculated by the difference between the current stability index and the baseline threshold; represents the cumulative amount of deformation factors of area i at historical time t; is the attenuation coefficient, used to adjust the influence weight of historical deformation factors; is the importance index of area i, the weight value standardized by the entropy weight method; is the base of the exponential function; is a natural constant, with a value of approximately 2.71828, is the exponential decay function, reflecting the cumulative amount of historical deformation factors has the characteristic of decaying over time for the influence on the current comprehensive reinforcement priority index .

[0046] Sort the reinforcement tasks of each area according to the comprehensive reinforcement priority index to generate a reinforcement task queue. The task sorting follows the principle of high to low priority, and the area at the head of the queue is reinforced first.

[0047] It should be noted that in the process of foundation pit soil reinforcement, setting the reinforcement priority based on the stability assessment and deformation prediction results is the key link to achieve efficient reinforcement and resource optimization. By quantifying the urgency of the reinforcement requirements for each monitoring area and combining the entropy weight method to determine the regional importance, the reinforcement resources can be allocated more scientifically. The urgency index is used to represent the intensity of the reinforcement requirements of the area at the current moment, while the regional importance index reflects the importance of the monitoring area in the overall structure of the foundation pit. The comprehensive reinforcement priority index is calculated by comprehensively considering the urgency and regional importance, and is used to sort the reinforcement tasks of each area to ensure that the key areas are reinforced first.

[0048] Specifically, the urgency index is calculated based on the difference between the current stability index and the baseline threshold, reflecting the gap between the current stability state of the soil mass and the safety standard. The cumulative deformation factor is the cumulative value of the deformation factor at historical time t, used to evaluate the cumulative degree of the soil mass deformation risk. The attenuation coefficient is used to adjust the influence weight of the historical deformation factor to ensure that in the calculation process, the deformation factor in the recent period has a greater impact on the urgency. The regional importance index is the weight value standardized by the entropy weight method, reflecting the relative importance of each monitoring area in the overall structure of the foundation pit. The entropy weight method is a weight calculation method based on information entropy. By constructing a regional importance judgment matrix and combining with the coefficient of variation method to calculate the weight distribution, the importance index of each area is finally obtained. The comprehensive reinforcement priority index is calculated by comprehensively considering the urgency index, the cumulative deformation factor, and the regional importance index, used to sort the reinforcement tasks of each area to ensure that the reinforcement resources can be preferentially allocated to the areas with the greatest need.

[0049] Preferably, when calculating the urgency index, a baseline threshold can be set, for example, 0.8, indicating that when the stability index is lower than this value, the area needs to be reinforced. The urgency index can be calculated by (baseline threshold - current stability index) / baseline threshold. The greater the difference, the higher the urgency. For the cumulative deformation factor, it can be calculated by accumulating the deformation factors at historical times and multiplying by the attenuation coefficient. The attenuation coefficient can be set to 0.9, indicating that every time unit passes, the influence weight of the historical deformation factor decreases by 10%. When constructing the regional importance judgment matrix, values can be assigned according to the spatial relationship between the area and the support structure. For example, the closer the area is to the support structure, the higher the assigned value. After calculating the weight distribution of the judgment matrix by the coefficient of variation method, the weight values are standardized to obtain the regional importance index of each area. Finally, the comprehensive reinforcement priority index can be obtained by weighted summation of the urgency index, the cumulative deformation factor, and the regional importance index. The weights can be adjusted according to the actual situation. For example, the weight of the urgency index is 0.4, the weight of the cumulative deformation factor is 0.3, and the weight of the regional importance index is 0.3. Sort the reinforcement tasks of each area according to the comprehensive reinforcement priority index, and the task sorting follows the principle of high to low priority to ensure that the reinforcement resources can be used efficiently and the areas with the highest risk are processed first.

[0050] In some embodiments, a multi-objective optimization algorithm is applied to evaluate the reinforcement resource requirements and dynamically monitor the material consumption, including the following steps: Evaluate the required amount of grouting material based on the priority and reinforcement amount of each reinforcement task; According to the reinforcement task type and soil properties, determine the curing agent ratio plan required for each task, and combine with the material inventory limit and supply capacity to calculate the material scheduling plan that meets the requirements of all tasks. For each reinforcement task, analyze the equipment requirements, including the model of the grouting machine and the drilling equipment, and calculate the duration of equipment occupancy; Based on the priority index of the task, material requirements, equipment requirements, and manpower requirements, and apply the ant colony optimization algorithm for dynamic resource scheduling to determine the resource scheduling cost.

[0051] It should be noted that during the foundation pit soil reinforcement process, applying the multi-objective optimization algorithm to evaluate the reinforcement resource requirements and dynamically monitor the material consumption is an important link to ensure the efficient and economical operation of the reinforcement project. The multi-objective optimization algorithm is an optimization method that can consider multiple objectives simultaneously, such as material consumption cost, reinforcement completion time, and resource utilization rate. By dynamically scheduling resources, including grouting materials, construction equipment, and monitoring personnel, the optimal allocation of resources can be achieved. This optimization method can generate a scientific and reasonable resource scheduling plan according to the priority, material requirements, equipment requirements, and manpower requirements of each reinforcement task, thereby improving the reinforcement efficiency and reducing the construction cost.

[0052] Specifically, the multi-objective optimization algorithm is a mathematical method for seeking the optimal solution among multiple objectives. In the present invention, the objective function includes material consumption cost, reinforcement completion time, and resource utilization rate. The material consumption cost refers to the cost of the grouting materials required to complete the reinforcement task; the reinforcement completion time refers to the total time required to complete all reinforcement tasks; the resource utilization rate refers to the utilization efficiency of resources such as grouting materials, construction equipment, and monitoring personnel. The resources to be scheduled include grouting materials, construction equipment, and monitoring personnel, and the demand for these resources is determined according to the scale and complexity of the reinforcement task. For example, the amount of grouting materials can be calculated based on the volume of the reinforcement area and the required grouting volume; the demand for construction equipment is determined according to the task type and equipment performance. The ant colony optimization algorithm is an optimization algorithm that simulates the foraging behavior of ants and finds the optimal path by simulating the pheromones released by ants during the process of finding food. In the present invention, the ant colony optimization algorithm is used for dynamic resource scheduling, and the reinforcement execution path is optimized by simulating the behavior of ants, thereby reducing the resource scheduling cost.

[0053] Preferably, the specific steps for constructing the multi-objective optimization model are as follows: First, evaluate the required amount of grouting material according to the priority and reinforcement amount of each reinforcement task. For example, the material requirement can be determined by calculating the volume of the reinforcement area and the unit consumption of the grouting material. Second, determine the curing agent ratio plan required for each task according to the type of reinforcement task and soil properties. The curing agent ratio plan can be adjusted according to factors such as the water content and particle size of the soil to ensure the reinforcement effect. Then, considering the material inventory limit and supply capacity, calculate the material scheduling plan that meets the requirements of all tasks. For example, if the inventory is limited, it is necessary to optimize the material allocation to ensure that high-priority tasks are given priority for materials. Next, analyze the equipment requirements for each reinforcement task, including the type of grouting machine and drilling equipment, and calculate the equipment occupancy time. The equipment requirements can be determined according to the scale and complexity of the task. For example, large-scale reinforcement tasks may require more powerful grouting equipment. Finally, apply the ant colony optimization algorithm for dynamic resource scheduling. The ant colony optimization algorithm simulates the behavior of ants to find the optimal resource allocation path. In the algorithm, each reinforcement task can be regarded as a node, and the quality of the resource allocation path is represented by the pheromone concentration. By continuously iterating and updating the pheromone concentration, the algorithm can find the optimal resource allocation scheme, thereby reducing the resource scheduling cost and improving the reinforcement efficiency.

[0054] In some embodiments, the use of the ant colony optimization algorithm to dynamically optimize the reinforcement scheme includes the following steps: Construct a multi-objective optimization model. Based on resource constraints and task priorities, construct a multi-objective optimization model. The objective function of the model includes material consumption cost, reinforcement completion time, and resource utilization rate. The resources to be scheduled include grouting materials, construction equipment, and monitoring personnel. Select the ant colony optimization algorithm to optimize the reinforcement execution path.

[0055] It should be noted that in the process of foundation pit soil reinforcement, the use of the ant colony optimization algorithm to dynamically optimize the reinforcement scheme is a key link to achieve efficient and economical reinforcement. The ant colony optimization algorithm is an optimization algorithm that simulates the foraging behavior of ants and finds the optimal path by simulating the pheromones released by ants during the process of finding food. In the present invention, this algorithm is used to dynamically optimize the reinforcement execution path, and can comprehensively consider multiple objectives such as material consumption cost, reinforcement completion time, and resource utilization rate under the conditions of resource constraints and task priorities, so as to generate the optimal reinforcement scheme and ensure the safety and economy of the foundation pit project.

[0056] Specifically, the ant colony optimization algorithm is an optimization method based on swarm intelligence, and its core is to simulate the behavior of ants in choosing paths by releasing pheromones during the process of finding food. In the present invention, the objective functions of the multi-objective optimization model include material consumption cost, reinforcement completion time, and resource utilization rate. The material consumption cost refers to the cost of grouting materials required to complete the reinforcement task; the reinforcement completion time refers to the total time required to complete all reinforcement tasks; the resource utilization rate refers to the utilization efficiency of resources such as grouting materials, construction equipment, and monitoring personnel. The resources to be scheduled include grouting materials, construction equipment, and monitoring personnel, and the demand for these resources is determined according to the scale and complexity of the reinforcement task. For example, the dosage of grouting materials can be calculated based on the volume of the reinforcement area and the required grouting volume; the demand for construction equipment is determined according to the task type and equipment performance. The ant colony optimization algorithm simulates the behavior of ants to find the optimal resource allocation path, thereby reducing the resource scheduling cost and improving the reinforcement efficiency.

[0057] Preferably, the specific steps for constructing the multi-objective optimization model are as follows: First, evaluate the required dosage of grouting materials according to the priority and reinforcement amount of each reinforcement task. For example, the material demand can be determined by calculating the volume of the reinforcement area and the unit dosage of grouting materials. Second, determine the curing agent mixing ratio plan required for each task according to the reinforcement task type and soil properties. The curing agent mixing ratio plan can be adjusted according to factors such as the water content and particle size of the soil to ensure the reinforcement effect. Then, considering the material inventory limit and supply capacity, calculate the material scheduling plan that meets the requirements of all tasks. For example, if the inventory is limited, it is necessary to optimize the material allocation to ensure that high-priority tasks are given priority for materials. Next, analyze the equipment requirements for each reinforcement task, including the type of grouting machine and drilling equipment, and calculate the equipment occupancy duration. The equipment requirements can be determined according to the scale and complexity of the task. For example, large-scale reinforcement tasks may require more powerful grouting equipment. Finally, apply the ant colony optimization algorithm for dynamic resource scheduling. The ant colony optimization algorithm simulates the behavior of ants to find the optimal resource allocation path. In the algorithm, each reinforcement task can be regarded as a node, and the quality of the resource allocation path is represented by the pheromone concentration. By continuously iterating and updating the pheromone concentration, the algorithm can find the optimal resource allocation plan, thereby reducing the resource scheduling cost and improving the reinforcement efficiency.

[0058] The above various embodiments of the present invention have the following beneficial effects: This method can collect the electrical parameter data of the foundation pit soil in real time, generate accurate soil state information through the adaptive weighted fusion algorithm, and automatically trigger re-collection when the data is abnormal to ensure the reliability of the monitoring data. The stability model constructed based on the fusion data can accurately evaluate the soil state. Combining the prediction function of the GRU model can detect potential instability risks in advance, and through the hierarchical early warning mechanism, corresponding measures can be taken in a timely manner to effectively prevent the occurrence of engineering accidents.

[0059] Through the entropy weight method and the multi-objective optimization algorithm, the reinforcement priority and resource allocation plan can be intelligently determined, enabling the maximum benefit to be achieved with limited construction resources. The application of the ant colony optimization algorithm can dynamically adjust the reinforcement path, significantly improving the reinforcement efficiency while ensuring the construction quality. The entire system can achieve full-process automatic control from data collection, risk warning to intelligent reinforcement, which can not only reduce the cost of manual intervention, but also improve the safety and economy of the foundation pit project.

[0060] Furthermore, the storage medium of the embodiment of the present application stores program instructions capable of implementing all the above methods. Among them, the program instructions can be stored in the above storage medium in the form of a software product, including several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) or a processor to execute all or part of the steps of the methods described in various embodiments of the present application. The aforementioned storage medium includes: various media that can store program codes such as USB flash drives, mobile hard disks, read-only memories (ROMs), random access memories (RAMs), magnetic disks, or optical discs, or terminal devices such as computers, servers, mobile phones, and tablets.

[0061] The above description is only some preferred embodiments of the present invention and an explanation of the applied technical principles. Those skilled in the art should understand that the scope of the invention involved in the embodiments of the present invention is not limited to the technical solutions formed by the specific combination of the above technical features, and should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the above inventive concept. For example, the technical solutions formed by mutually replacing the above features with the (but not limited to) technical features with similar functions disclosed in the embodiments of the present invention.

Claims

1. A method for reinforcing foundation pit soil based on electrical parameter monitoring, characterized in that, It includes the following steps: Embed an electrode array in the foundation pit soil body to collect electrical parameter data, perform spatio-temporal alignment processing on the data, generate soil body state data through an adaptive weighted fusion algorithm for the aligned data, and trigger re-collection and calibration of the data when the fluctuation exceeds the threshold; Construct a soil body stability model based on the fusion data, calculate the stability index to evaluate the soil body state, use moving average filtering to smooth short-term fluctuations, predict future stability changes through the GRU model, trigger an early warning when it is below the safety baseline, divide the risk level according to the stability index and deformation factor, and generate a reinforcement instruction when it is at high risk; Set the reinforcement priority based on stability assessment and deformation prediction, calculate the urgency by combining the stability index and deformation risk, determine the regional importance through the entropy weight method, evaluate the reinforcement resource requirements using a multi-objective optimization algorithm, and dynamically monitor the material consumption situation; Use the ant colony optimization algorithm to dynamically optimize the reinforcement plan.

2. The method for reinforcing foundation pit soil based on electrical parameter monitoring according to claim 1, characterized in that, Embed an electrode array in the foundation pit soil body to collect electrical parameter data, perform spatio-temporal alignment processing on the data, generate soil body state data through an adaptive weighted fusion algorithm for the aligned data, and trigger re-collection and calibration of the data when the fluctuation exceeds the threshold. The specific steps are as follows: The data collected by the electrode array includes resistivity parameter, capacitance value parameter, and dielectric constant parameter; Perform normalization processing on the collected data to make the data ranges of different electrical parameters consistent; Use the LOF local outlier factor algorithm to perform anomaly detection on each normalized data, remove noise data, and perform spatio-temporal alignment processing on the collected data to align data at different spatial positions to the same time series; After completing the spatio-temporal alignment, perform fusion on the collected data based on the adaptive weighted fusion algorithm with dynamic weight allocation, and the weight allocation is dynamically adjusted according to the historical correlation between each electrical parameter and the soil body stability; The fused data is used as the input of the soil body state at the current time point, and a feedback mechanism is set. The feedback mechanism is that when the difference between the fusion value and the value at the previous time point exceeds the set threshold, restart the data collection of the electrode array and calibrate the measuring device.

3. The method for reinforcing foundation pit soil based on electrical parameter monitoring according to claim 2, characterized in that, Construct a soil body stability model based on the fusion data, calculate the stability index to evaluate the soil body state, use moving average filtering to smooth short-term fluctuations. The specific steps include: The fusion data is used as the input data for stability assessment. Establish a soil body stability model for the fusion data, calculate the stability index of the soil body through multiple regression analysis and the generation of covariance matrix, and determine the stable state of the soil body according to the stability index; Use moving average filtering to dynamically smooth the short-term fluctuation impact on the stability index of the soil body; For the area where the stable state is at the critical value, extract the deviation degree of the characteristic value of the fusion data from the safety threshold to determine the deformation factor, and use it to analyze the structural risk.

4. The method for reinforcing foundation pit soil based on electrical parameter monitoring according to claim 3, characterized in that, Predict future stability changes through the GRU model, and trigger an early warning when it is below the safety baseline. The specific steps are as follows: Construct a deformation prediction model based on historical data, and use the GRU model to predict the stability index at future moments; The input of the deformation prediction model is the current and historical stability indices and deformation factors; If the predicted stability index continues to be lower than the safety baseline threshold, trigger an early warning signal to indicate the risk of soil body instability.

5. The method for reinforcing foundation pit soil based on electrical parameter monitoring according to claim 4, characterized in that, Divide the risk levels according to the stability index and the deformation factor, and generate reinforcement instructions in the case of high risk, including the following steps: Determine the risk level of each monitoring area according to the decline rate of the stability index and the cumulative amount of the deformation factor. The risk levels include low risk level, warning risk level, and high risk level; When the risk level is the low risk level, no reinforcement instruction is generated; When the risk level is the warning risk level, mark the monitoring area as a warning area and formulate a monitoring enhancement plan; When the risk level is the high risk level, generate a reinforcement instruction signal to trigger the automatic grouting system for active reinforcement; Among them, the conditions for the low risk level are: The decline rate is lower than the first decline rate threshold, and the cumulative amount of the deformation factor is lower than the first deformation factor cumulative amount threshold; The conditions for the warning risk level are: The decline rate reaches or exceeds the first decline rate threshold but is lower than the second decline rate threshold, and the cumulative amount of the deformation factor is lower than the first deformation factor cumulative amount threshold; or, the cumulative amount of the deformation factor reaches or exceeds the first deformation factor cumulative amount threshold but is lower than the second deformation factor cumulative amount threshold, and the decline rate is lower than the first decline rate threshold; The conditions for the high risk level are: The decline rate reaches or exceeds the second decline rate threshold; or, the cumulative amount of the deformation factor reaches or exceeds the second deformation factor cumulative amount threshold; or, the decline rate reaches or exceeds the first decline rate threshold and the cumulative amount of the deformation factor reaches or exceeds the first deformation factor cumulative amount threshold.

6. The method for reinforcing foundation pit soil based on electrical parameter monitoring according to claim 5, characterized in that,Set the reinforcement priority based on stability assessment and deformation prediction, calculate the urgency in combination with the stability index and deformation risk, and determine the regional importance through the entropy weight method, including the following steps: Quantify the urgency of the reinforcement requirements for each monitoring area according to the stability assessment and deformation prediction results. The urgency index is used to represent the intensity of the reinforcement requirements in the area at the current moment; Determine the relative importance of each monitoring area through the entropy weight method, construct a regional importance judgment matrix, and assign values according to the spatial relationship between the area and the support structure; Calculate the weight distribution of the judgment matrix by the coefficient of variation method, standardize the calculation results, and obtain the weight of each area as the regional importance index; After the calculation is completed, determine the key reinforcement area according to the obtained regional importance index value; After obtaining the task urgency and regional importance, calculate the comprehensive reinforcement priority index. The expression formula is: Among them, is the comprehensive reinforcement priority index of area i; is the adjustment parameter; is the urgency index of area i; represents the cumulative amount of deformation factors of area i at historical moment t; is the attenuation coefficient; is the importance index of area i; is the base of the exponential function; Sort the reinforcement tasks of each area according to the comprehensive reinforcement priority index to generate a reinforcement task queue. The task sorting follows the principle of high to low priority, and the area at the head of the queue is reinforced first.

7. A method for reinforcing foundation pit soil based on electrical parameter monitoring according to claim 6, characterized in that, Apply a multi-objective optimization algorithm to evaluate the reinforcement resource requirements and dynamically monitor the material consumption, including the following steps: Evaluate the required amount of grouting material based on the priority and reinforcement volume of each reinforcement task; Determine the curing agent mixing ratio plan required for each task according to the reinforcement task type and soil properties, and calculate the material scheduling plan that meets all task requirements in combination with the material inventory limit and supply capacity; For each reinforcement task, analyze the equipment requirements, including the grouting machine model and drilling equipment, and calculate the equipment occupation duration; Based on the task priority index, material requirements, equipment requirements, and manpower requirements, and applying the ant colony optimization algorithm for dynamic resource scheduling to determine the resource scheduling cost.

8. A method for reinforcing foundation pit soil based on electrical parameter monitoring according to claim 7, characterized in that, Use the ant colony optimization algorithm to dynamically optimize the reinforcement plan, including the following steps: Construct a multi-objective optimization model. Based on resource constraints and task priorities, construct a multi-objective optimization model. The objective function of the model includes material consumption cost, reinforcement completion time, and resource utilization rate. The resources to be scheduled include grouting materials, construction equipment, and monitoring personnel. Select the ant colony optimization algorithm to optimize the reinforcement execution path.

9. An electronic device, comprising: At least one processor; And a memory communicatively connected to the at least one processor; wherein, the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the method according to any one of claims 1-8.

10. A computer-readable storage medium, characterized in that, The storage medium stores computer instructions. When the computer reads the computer instructions in the storage medium, the computer executes the method according to any one of claims 1-8.

Images