Surrounding environment risk assessment method based on cover excavation subway foundation pit

Abstract

The invention discloses a surrounding environment risk assessment method based on a cover excavation subway foundation pit, and relates to the technical field of subway construction. The evaluation method comprises the following steps of multi-source data acquisition and preprocessing: collecting geological survey data, environment monitoring data, construction monitoring data, existing building data, pipeline data and electric power facility data of a foundation pit area. The deep analysis of the complex interaction between the foundation pit construction and the surrounding environment is realized, the innovative method overcomes the limitations of single data and one-sided analysis in the traditional risk assessment, the comprehensiveness and accuracy of the risk assessment are remarkably improved, and the risk assessment efficiency is improved through comprehensive analysis. The potential influence of foundation pit construction on the surrounding environment can be accurately monitored in real time, the reasonable early warning threshold value is set, and once the monitoring data exceed the preset threshold value, the early warning signal is triggered immediately, so that the safety of the surrounding environment is effectively guaranteed, and the risk in the construction process is reduced.

Description

Technical Field

[0001] The invention relates to the technical field of subway construction, and in particular to a surrounding environment risk assessment method based on covering and excavating a subway foundation pit. Background Art

[0002] With the acceleration of urbanization, subway construction, as an important means to alleviate urban traffic pressure, has become increasingly important. However, subway foundation pit construction is characterized by large engineering volume, long construction period and complex technology. The impact on the surrounding environment during construction has become a difficult problem to solve. Traditional risk assessment methods for the surrounding environment of foundation pits mainly rely on a single data source and simple analysis methods, which are often difficult to fully and accurately reflect the complex interaction between foundation pit construction and the surrounding environment. This limitation leads to the one-sidedness and inaccuracy of risk assessment results, which makes it difficult to provide a strong scientific basis for risk prevention and control.

[0003] Traditional risk assessment methods mainly rely on empirical judgment, monitoring of a single data source, and simple statistical analysis. These methods are powerless when dealing with the complex interaction between foundation pit construction and the surrounding environment. On the one hand, a single data source often cannot fully reflect the multi-dimensional impact of foundation pit construction on the surrounding environment. On the other hand, the lack of advanced analytical methods makes risk assessment results susceptible to subjective judgment and one-sided information, resulting in inaccurate and unreliable assessment results. In addition, traditional methods lack real-time and dynamic characteristics, making it difficult to respond to changes in the construction process in a timely manner and unable to provide timely and effective support for risk prevention and control. Traditional risk assessment technology has obvious limitations and shortcomings when facing the complex system of foundation pit construction.

[0004] Therefore, in view of the deficiencies and shortcomings of the prior art, the present invention proposes a surrounding environment risk assessment method based on covered and excavated subway foundation pits, which greatly improves the accuracy and pertinence of risk assessment, provides strong support for the formulation of scientific and reasonable preventive measures, effectively ensures construction safety and effective protection of the surrounding environment, and enhances the social and environmental benefits of construction projects. Summary of the invention

[0005] The purpose of the present invention is to make up for the shortcomings of the prior art and to provide a surrounding environment risk assessment method based on a covered and excavated subway foundation pit. It realizes refined risk management of the entire foundation pit construction process through multi-source data collection and preprocessing, data fusion, comprehensive analysis, risk assessment and early warning, and continuous monitoring and optimization steps. The method deeply reveals the complex relationship between foundation pit construction and the surrounding environment by using data mining and machine learning algorithms, providing comprehensive and accurate data support for risk assessment.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions: a surrounding environment risk assessment method based on the covered excavation of a subway foundation pit, the specific steps of the assessment method are:

[0007] S100, multi-source data collection and preprocessing: collect geological survey data, environmental monitoring data, construction monitoring data, existing building data, pipeline data and power facility data of the foundation pit area, and preprocess the collected multi-source data;

[0008] S200, multi-source data fusion, synchronizes time and aligns space of pre-processed multi-source data to construct a comprehensive dataset of the interaction between foundation pit construction and the surrounding environment;

[0009] S300, comprehensive analysis: Use machine learning algorithms to conduct comprehensive analysis on the fused multi-source data to determine the relationship between foundation pit construction parameters and the settlement of surrounding existing buildings, pipeline deformation, and the operating status of power facilities;

[0010] S400, Risk Assessment and Early Warning: Based on the results of the comprehensive analysis model, conduct a quantitative risk assessment of the impact on surrounding existing buildings, pipelines and power facilities, calculate the risk level and probability of occurrence, and set up a risk early warning mechanism;

[0011] S500, continuous monitoring and optimization: Continuously monitor changes in multi-source data during the construction process, and continuously optimize risk assessment models and control measures based on new data and actual conditions.

[0012] Further, the S100 includes:

[0013] S101, using geological exploration methods to obtain detailed geological survey data of the foundation pit area, using environmental monitoring equipment to regularly collect environmental parameters of the foundation pit area and its surroundings, using the monitoring system of the construction site to collect various monitoring data during the construction process in real time, and using the geographic information system platform to integrate and analyze existing building data, pipeline data and power facility data;

[0014] S102, performing integrity check on the collected data and removing noise and outliers;

[0015] S103, converting all collected multi-source data into a unified format and unit.

[0016] Furthermore, the S200 includes:

[0017] S201, determine a unified time standard as a benchmark, check and convert the timestamps of each data source, perform time alignment processing on non-real-time data sources and data sources with different acquisition frequencies, and divide the time axis into several time windows according to analysis requirements;

[0018] S202, convert all data sources into the same spatial coordinate system, clarify the spatial range required for analysis, match data sources with different spatial resolutions, and use GIS technology to establish the spatial relationship between foundation pit construction and the surrounding environment;

[0019] S203, integrating the time-synchronized and space-aligned multi-source data into a unified comprehensive data set, fusing the multi-source data in the comprehensive data set, and verifying the fused data.

[0020] Furthermore, the S300 includes:

[0021] S301, selecting features that have an impact on the risk assessment of the surrounding environment of foundation pit construction from the fused data set;

[0022] S302, using support vector machine to analyze the correlation between foundation pit deformation and building cracks;

[0023] S303, using random forest to predict the squeezing and stretching effects of soil displacement caused by foundation pit excavation on pipelines, and the risk of pipeline rupture;

[0024] S304, using a neural network to analyze interference to the power lines during the construction process.

[0025] Furthermore, the S302 uses a support vector machine to analyze the correlation between foundation pit deformation and building cracks, and selects the RBF kernel function as the support vector machine kernel function, and the formula is: Among them, x and x′ are two input vectors, σ is the width parameter of the kernel function, and the goal of SVM is to find a hyperplane that maximizes the interval between data points of different categories. For predicting the correlation between foundation pit deformation and building cracks, the objective function formula is: The constraints are: Among them, w is the weight vector, b is the bias term, and y i is the crack width, x i is the pit deformation parameter, ∈ is the tolerance error, C is the regularization parameter, ξ i and is a slack variable. The data is divided into training set, validation set and test set. The training set is used to train the SVM model to find the optimal w, b, The performance of the model was evaluated using the validation set and the test set. The model parameters were adjusted based on the evaluation results. The trained SVM model was applied to actual scenarios to predict the correlation between foundation pit deformation and building cracks, and corresponding risk prevention and control measures were taken based on the prediction results.

[0026] Furthermore, the 303 uses a random forest to predict the risk of pipeline rupture caused by soil displacement during foundation pit excavation. The random forest represents the potential rupture risk of the pipeline with a binary variable R. When R = 0, it indicates no rupture risk, and when R = 1, it indicates a rupture risk. For each decision tree, m samples are randomly sampled with replacement from the original dataset to form a new training set for constructing the decision tree. When splitting at each node, k features are randomly selected from p feature variables, where p < k. The best splitting feature and splitting point are selected according to the Gini index. The formula for the Gini index is: where K is divided into two categories, rupture risk and no rupture risk, when predicting the potential rupture risk of the pipeline, and p k is p0 and p1. p0 represents the probability that the sample belongs to the category of no rupture risk, and p1 represents the probability that the sample belongs to the category of rupture risk. Repeat the above steps to construct T decision trees to form a random forest. For the prediction of the potential rupture risk R of the pipeline, the predicted value of the potential rupture risk of the pipeline is determined by the majority voting method. Count the number of R = 0 and R = 1 in all the prediction results of the decision trees, and take the category with the larger number as the final prediction category.

[0027] Furthermore, the 304 uses a neural network to analyze the interference caused to the power line during the construction process. The neural network consists of an input layer, a hidden layer, and an output layer. Each layer contains multiple neurons, and the neurons are interconnected through weights and biases. The sigmoid function is used as the activation function in the neurons to introduce non-linear factors. The sigmoid function is as follows: where z is the input of the neuron. In the neural network, the input data propagates forward through the neurons of each layer and finally obtains the output. For the j-th neuron in the l-th layer, its output is expressed as: where n (l-1) is the number of neurons in the l-1-th layer, w j k (l) is the weight from the k-th neuron in the l-1-th layer to the j-th neuron in the l-th layer, is the bias of the j-th neuron in the l-th layer, and f(·) is the activation function. The difference between the predicted value and the actual value is measured by calculating the mean square error. The output of the output layer is calculated as: where is the output of the j-th neuron in the hidden layer, is the weight from the hidden layer to the output layer, is the bias of the neuron in the output layer, and M is the number of neurons in the hidden layer. The mean square error (MSE) is used as the loss function, and the formula is: where N is the number of samples, and y iis the actual value of the ith sample, is the predicted value of the i-th sample. During the training process, the gradient of the loss function with respect to each weight and bias is calculated through the back propagation algorithm, and the gradient descent method is used to update these parameters to minimize the loss function. The performance of the model is evaluated using a data set that did not participate in the training. The trained neural network model is applied to actual scenarios to predict the interference to the power lines during the construction process, and corresponding measures are taken to reduce the interference based on the prediction results.

[0028] Furthermore, the S400 includes:

[0029] S401, comprehensively evaluate the impact of foundation pit excavation on the safety of surrounding buildings, water supply pipelines, and power facilities, and construct an assessment model based on the assessment objectives and data characteristics;

[0030] S402, based on data and expert opinions, identify the main risk factors, quantify the risk factors using linear regression, determine their probability of occurrence and impact, analyze the interaction and correlation between factors, calculate the risk level and probability of occurrence of each risk factor, combine all risk factors to obtain the overall risk level, and divide it into three risk levels: low, medium, and high;

[0031] S403, according to the risk assessment results and actual situation, set reasonable warning thresholds for different risk levels and risk factors, and formulate corresponding warning strategies for each warning threshold;

[0032] S404: Once the monitoring data exceeds the preset warning threshold, the warning signal is immediately triggered, and notification and response are carried out according to the warning strategy.

[0033] Furthermore, the S402 uses linear regression to quantify the risk factors to determine their occurrence probability and impact, and identifies the main risk factors that affect the safety of surrounding buildings, pipelines and power facilities as independent variables X, Y i Represents the value of the dependent variable. The general form of the linear regression model is: Y = β0 + β1X1 + β2X2 + ... + β n X n +∈, where β0 is the intercept, which represents the value of the dependent variable Y when all independent variables are 0, β1,β2,…,β n is the regression coefficient, which indicates the influence of each independent variable on the dependent variable. ∈ is the error term, which represents the part that cannot be explained by the model. The least squares method is used to estimate the model parameters. According to the obtained regression coefficient The absolute value of the regression coefficient reflects the degree of influence of the risk factor on the dependent variable. The established linear regression model is used to predict the probability of occurrence and the degree of influence under new circumstances.

[0034] Furthermore, the S500 includes:

[0035] S501, collect various data from the construction site in real time, clean, integrate and pre-process the collected data, conduct in-depth analysis of the processed data, and identify potential risks and problems;

[0036] S502: Dynamically adjust the existing risk assessment model based on new data and actual conditions, verify the newly adjusted risk assessment model, and formulate targeted control measures;

[0037] S503, regularly evaluate the effectiveness of the implementation of control measures, and continuously optimize the monitoring and optimization process based on the evaluation results and feedback.

[0038] Compared with the prior art, the surrounding environment risk assessment method based on the covered and excavated subway foundation pit in the present invention has the following beneficial effects:

[0039] 1. The present invention realizes an in-depth analysis of the complex interaction between foundation pit construction and the surrounding environment by comprehensively using data mining and machine learning algorithms, overcomes the limitations of single data and one-sided analysis in traditional risk assessment, and significantly improves the comprehensiveness and accuracy of risk assessment. Through comprehensive analysis, it can monitor the potential impact of foundation pit construction on the surrounding environment in real time and accurately, and set a reasonable early warning threshold. Once the monitoring data exceeds the preset threshold, the early warning signal is immediately triggered, which effectively ensures the safety of the surrounding environment and reduces the risks in the construction process.

[0040] 2. Through comprehensive analysis of data mining and machine learning algorithms, the present invention can deeply reveal how various factors interact during foundation pit construction and thus affect the surrounding environment. This innovation not only improves the accuracy of risk assessment, but also provides a scientific basis for risk prevention and control. Through quantitative risk assessment and probability calculation, it can accurately judge the risk level and occurrence probability of different risk factors, and provide strong support for decision makers to formulate targeted prevention and control measures. In addition, the continuous monitoring and optimization mechanism ensures the dynamic adjustment of the risk assessment model, so that it can better adapt to changes in actual construction conditions, further improving the effectiveness and effectiveness of risk prevention and control.

[0041] Other advantages, objectives and features of the present invention will be set forth in part in the following description and, in part, will be apparent to those skilled in the art based on an examination of the following or may be taught from the practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention, and for ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0043] Figure 1 It is a flowchart framework of a surrounding environment risk assessment method based on covered excavation subway foundation pit;

[0044] Figure 2 The figure is a flow chart of a surrounding environment risk assessment method based on covered and excavated subway foundation pits. DETAILED DESCRIPTION

[0045] The technical solutions in the embodiments of the present invention are described clearly and completely below. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0046] Embodiment 1:

[0047] A method for risk assessment of the surrounding environment based on a covered and excavated subway foundation pit comprises the following steps: multi-source data collection and preprocessing, multi-source data fusion, comprehensive analysis, risk assessment and early warning, and continuous monitoring and optimization.

[0048] First, enter the multi-source data collection and preprocessing stage (S100), use geological exploration technology to conduct detailed geological exploration of the foundation pit area, obtain data on soil layer distribution, groundwater level, rock type and strength, install environmental monitoring stations in the foundation pit area and its surroundings to collect environmental parameters, deploy high-definition cameras and intelligent sensors around and inside the foundation pit, and collect key monitoring data during the construction process in real time. Use the geographic information system platform to integrate existing building data, pipeline data, and power facility data, perform integrity verification on various types of collected data, identify and eliminate noise and outliers in the data, and convert all collected multi-source data into a unified file format and data structure.

[0049] Then, enter the multi-source data fusion analysis stage (S200), select UTC as the only time standard for all data sources, review the timestamp of each data source and convert it into UTC time, use interpolation and aggregation methods to align on the time axis for non-real-time data and data sources with different acquisition frequencies, divide the time axis into several time windows according to analysis requirements, select WGS84 as the unified spatial coordinate system for all data sources, convert all data sources from their respective coordinate systems to the WGS84 coordinate system, clarify the spatial range required for analysis, match data sources with different spatial resolutions to ensure that they can correspond to each other in space, use GIS technology to establish the spatial relationship between foundation pit construction and the surrounding environment, mark the locations of key elements of existing buildings, pipelines, and power facilities, integrate the multi-source data after time synchronization and spatial alignment into a unified comprehensive data set, and fuse the multi-source data in the comprehensive data set to extract higher-level information and features.

[0050] Subsequently, the comprehensive analysis stage (S300) is entered. From the fused data set, features directly related to the risk assessment of the surrounding environment of the foundation pit construction are screened out, including the amount of foundation pit deformation, the width and length of building cracks, the material type and burial depth of pipelines, and the voltage level and location of power lines. The selected features are further processed to extract higher-level features and indicators, and the foundation pit deformation data and building crack data are sorted out. The support vector machine (SVM) algorithm is used to construct a correlation model between foundation pit deformation and building cracks. The input features include the amount of foundation pit deformation, deformation rate, and deformation direction, and the output is the change of building cracks. The process of foundation pit excavation is sorted out. The soil displacement data, pipeline material property data and pipeline burial condition data are used, and the random forest algorithm is used to build a model to predict the squeezing and stretching effects of soil displacement caused by foundation pit excavation on pipelines. The input features include soil displacement amount, displacement direction, pipeline material properties, and burial conditions. The output is the probability of pipeline rupture risk. The monitoring data of power lines during construction and the location and operating status data of construction machinery are sorted out, and a neural network is used to build a model to analyze the interference to power lines during construction. The input features include the location and operating status of construction machinery, and the voltage and current data of power lines. The output is the degree of interference and stability assessment of power lines.

[0051] On the basis of the above, we enter the risk assessment and early warning stage (S400), comprehensively assess the impact of foundation pit excavation on the safety of surrounding buildings, water supply pipelines, and power facilities, build an assessment model based on the fault tree analysis method, identify the main risk factors based on data and expert opinions, and use a linear regression model. The general form of the linear regression model is: Y = β0 + β1X1 + β2X2 + ... + β n X n+∈, use the least squares method to estimate the model parameters, based on the obtained regression coefficients is positive, indicating that the independent variable X j The increase in will lead to an increase in the dependent variable Y. is negative, which means X j The increase of will reduce Y. The absolute value of the regression coefficient reflects the degree of influence of the risk factor on the dependent variable. Combining historical data with on-site monitoring data, the probability of occurrence and the degree of influence of each risk factor are quantified, and the interaction and correlation between factors are analyzed. According to the probability of risk occurrence and the degree of influence, the risk matrix method is used to calculate the risk level of each risk factor. All risk factors are combined to obtain the overall risk level, which is divided into three risk levels: low, medium and high. Reasonable early warning thresholds are set for different risk levels and risk factors. For low-risk factors, routine monitoring and recording are adopted. For medium-risk factors, enhanced monitoring plans are initiated and emergency plans are prepared. For high-risk factors, construction is stopped immediately, emergency response mechanisms are initiated, surrounding residents are evacuated, and relevant departments are contacted for assistance. Key parameters of the foundation pit and surrounding environment are monitored in real time. Once the monitoring data exceeds the preset early warning threshold, the early warning signal is automatically triggered.

[0052] Finally, enter the continuous monitoring and optimization stage (S500), collect environmental data, construction equipment status data, and personnel activity data from the construction site in real time, conduct preliminary screening of the collected data, remove duplications, errors, and outliers, standardize the data set, dynamically adjust the existing risk assessment model based on the newly collected data and actual conditions, use historical data and some newly collected data to verify the adjusted model, formulate targeted control measures based on the results of the risk assessment model, regularly evaluate the implementation effect of the control measures, and continuously optimize the monitoring and optimization process based on the evaluation results and feedback. Treat evaluation and optimization as a continuous process, continuously collect new data, update the risk assessment model, adjust control measures, and evaluate their effectiveness.

[0053] Embodiment 2:

[0054] A method for risk assessment of the surrounding environment based on the excavation and cover of subway foundation pits can fully understand the impact of foundation pit construction on the surrounding environment through comprehensive analysis:

[0055] Collect foundation pit deformation monitoring data, building crack monitoring data, pipeline status data, and power line status data, fuse the multi-source data to form a unified data set, and select from the fused data set features that are meaningful for the risk assessment of the surrounding environment of foundation pit construction, such as foundation pit deformation, deformation rate, building crack width, crack propagation speed, pipeline material properties, burial depth, power line voltage fluctuations, and current changes.

[0056] Support vector machine is used to analyze the correlation between foundation pit deformation and building cracks. RBF kernel function is selected as the support vector kernel function. The RBF kernel function formula is: The goal of SVM is to find a hyperplane that maximizes the interval between data points of different categories. For predicting the correlation between foundation pit deformation and building cracks, the objective function formula is: Constraints: The data are divided into training set, validation set and test set. The training set is used to train the SVM model to find the optimal,w,b,∈,C,. The validation set and test set are used to evaluate the model performance and adjust the model parameters. The trained SVM model is applied to the actual scenario to predict the correlation between foundation pit deformation and building cracks, and corresponding risk prevention and control measures are taken according to the prediction results.

[0057] Random forest is used to predict the risk of pipeline rupture caused by soil displacement caused by foundation pit excavation. The potential rupture risk of the pipeline is represented by a binary variable R. When R = 0, it means there is no rupture risk, and when R = 1, it means there is a rupture risk. For each decision tree, m samples are extracted from the original data set using random sampling with replacement to form a new training set to construct a decision tree. When splitting at each node, k features are randomly selected from p feature variables. The best split features and split points are selected according to the Gini index. The Gini index calculation formula is: The new sample X new Input to each decision tree in the random forest to obtain the prediction value R of each tree for the potential risk of pipeline rupture t , the predicted value of the potential risk of pipeline rupture The majority voting method is used to determine the number of R=0 and R=1 in all decision tree prediction results, and the category with the largest number is taken as the final prediction category.

[0058] The neural network is used to analyze the interference caused to the power lines during the construction process, collect the status data of the power lines during the construction process, and the data of the factors that cause the interference of the power lines. The collected data is cleaned to remove noise and outliers, and the data is divided into training set, verification set and test set. The neural network consists of an input layer, a hidden layer and an output layer. Each layer contains multiple neurons. The neurons are connected to each other through weights and biases. The sigmoid function is used as an activation function in the neurons to introduce nonlinear factors. The sigmoid function is used as: In a neural network, input data propagates forward through each layer of neurons and finally gets the output. For the jth neuron in the lth layer, its output is It is expressed as: The output of the output layer is calculated as: The mean square error (MSE) is used as the loss function, and the formula is: The gradient of the loss function with respect to each weight and bias is calculated through the back-propagation algorithm, and the weights and biases are updated using the gradient descent method to minimize the loss function. The learning rate is adjusted to control the step size of the parameter update, and the network architecture is adjusted according to the performance of the validation set. The performance of the model is evaluated using the test set to ensure that the model has good generalization ability. The trained neural network model is deployed to the actual construction site, and the data input during the construction process is received in real time. The model outputs the predicted degree of power line interference, and corresponding measures are taken in a timely manner according to the prediction results to reduce interference to the power lines.

[0059] It will be apparent to those skilled in the art that the invention is not limited to the details of the exemplary embodiments described above and that the invention can be implemented in other specific forms without departing from the spirit or essential features of the invention. Therefore, the embodiments should be considered exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description, and it is intended that all variations falling within the meaning and scope of the equivalent elements of the claims be included in the invention. Any reference numeral in a claim should not be considered as limiting the claim to which it relates.

Claims

1. A method for risk assessment of surrounding environment based on covered and excavated subway foundation pit, characterized in that: The specific steps of this evaluation method are: S100, multi-source data collection and preprocessing: collect geological survey data, environmental monitoring data, construction monitoring data, existing building data, pipeline data and power facility data of the foundation pit area, and preprocess the collected multi-source data; S200, multi-source data fusion, synchronizes time and aligns space of pre-processed multi-source data to construct a comprehensive dataset of the interaction between foundation pit construction and the surrounding environment; S300, comprehensive analysis: Use machine learning algorithms to conduct comprehensive analysis on the fused multi-source data to determine the relationship between foundation pit construction parameters and the settlement of surrounding existing buildings, pipeline deformation, and the operating status of power facilities; S400, Risk Assessment and Early Warning: Based on the results of the comprehensive analysis model, conduct a quantitative risk assessment of the impact on surrounding existing buildings, pipelines and power facilities, calculate the risk level and probability of occurrence, and set up a risk early warning mechanism; S500, continuous monitoring and optimization: Continuously monitor changes in multi-source data during the construction process, and continuously optimize risk assessment models and control measures based on new data and actual conditions.

2. According to claim 1, a method for risk assessment of surrounding environment based on covering and excavating subway foundation pits is characterized in that: The S100 includes: S101, using geological exploration methods to obtain detailed geological survey data of the foundation pit area, using environmental monitoring equipment to regularly collect environmental parameters of the foundation pit area and its surroundings, using the monitoring system of the construction site to collect various monitoring data during the construction process in real time, and using the geographic information system platform to integrate and analyze existing building data, pipeline data and power facility data; S102, performing integrity check on the collected data and removing noise and outliers; S103, converting all collected multi-source data into a unified format and unit.

3. According to claim 1, a method for risk assessment of surrounding environment based on covering and excavating subway foundation pits is characterized in that: The S200 includes: S201, determine a unified time standard as a benchmark, check and convert the timestamps of each data source, perform time alignment processing on non-real-time data sources and data sources with different acquisition frequencies, and divide the time axis into several time windows according to analysis requirements; S202, convert all data sources into the same spatial coordinate system, clarify the spatial range required for analysis, match data sources with different spatial resolutions, and use GIS technology to establish the spatial relationship between foundation pit construction and the surrounding environment; S203, integrating the time-synchronized and space-aligned multi-source data into a unified comprehensive data set, fusing the multi-source data in the comprehensive data set, and verifying the fused data.

4. According to claim 1, a method for risk assessment of surrounding environment based on covering and excavating subway foundation pits is characterized in that: The S300 includes: S301, selecting features that have an impact on the risk assessment of the surrounding environment of foundation pit construction from the fused data set; S302, using support vector machine to analyze the correlation between foundation pit deformation and building cracks; S303, using random forest to predict the squeezing and stretching effects of soil displacement caused by foundation pit excavation on pipelines, and the risk of pipeline rupture; S304, using a neural network to analyze interference to the power lines during the construction process.

5. The method for risk assessment of surrounding environment based on covering and excavating subway foundation pit according to claim 4 is characterized in that: The S302 uses a support vector machine to analyze the correlation between foundation pit deformation and building cracks, and selects the RBF kernel function as the support vector machine kernel function, and the formula is: Among them, x and x ′ are two input vectors, σ is the width parameter of the kernel function, and the goal of SVM is to find a hyperplane that maximizes the interval between data points of different categories. For predicting the correlation between foundation pit deformation and building cracks, the objective function formula is: The constraints are: Among them, w is the weight vector, b is the bias term, and y i is the crack width, x i is the pit deformation parameter, ∈ is the tolerance error, C is the regularization parameter, ξ i and is a slack variable. The data is divided into training set, validation set and test set. The training set is used to train the SVM model to find the optimal w, b, The performance of the model was evaluated using the validation set and the test set. The model parameters were adjusted based on the evaluation results. The trained SVM model was applied to actual scenarios to predict the correlation between foundation pit deformation and building cracks, and corresponding risk prevention and control measures were taken based on the prediction results.

6. The method for risk assessment of surrounding environment based on covering and excavating subway foundation pit according to claim 4 is characterized in that: The 303 uses random forest to predict the risk of pipeline rupture caused by soil displacement during foundation pit excavation. The potential rupture risk of the pipeline is represented by a binary variable R. When R = 0, it indicates no rupture risk, and when R = 1, it indicates a rupture risk. For each decision tree, m samples are randomly sampled with replacement from the original dataset to form a new training set for constructing the decision tree. When splitting at each node, k features are randomly selected from p feature variables, where p < k. The best splitting feature and splitting point are selected according to the Gini index. The formula for the Gini index is: where K: When predicting the potential rupture risk of the pipeline, it is divided into two categories: rupture risk and no rupture risk, p k are p0 and p1. p0 represents the probability that the sample belongs to the category of no rupture risk, and p1 represents the probability that the sample belongs to the category of rupture risk. Repeat the above steps to construct T decision trees to form a random forest. For the prediction of the potential rupture risk R of the pipeline, the predicted value of the potential rupture risk of the pipeline is determined by the majority voting method. Count the number of R = 0 and R = 1 in the prediction results of all decision trees, and take the category with the larger number as the final prediction category.

7. The method for risk assessment of surrounding environment based on covering and excavating subway foundation pit according to claim 4 is characterized in that: The 304 uses a neural network to analyze the interference to the power line during the construction process. The neural network consists of an input layer, a hidden layer, and an output layer. Each layer contains multiple neurons. The neurons are interconnected through weights and biases. The sigmoid function is used as an activation function in the neurons to introduce nonlinear factors. The sigmoid function is used as: Among them, z is the input of the neuron. In the neural network, the input data propagates forward through each layer of neurons and finally gets the output. For the jth neuron in the lth layer, its output is It is expressed as: Among them, n (l-1) is the number of neurons in the l-1th layer, w j k (1) is the weight from the kth neuron in the l-1th layer to the jth neuron in the lth layer, is the bias of the jth neuron in the lth layer, f(·) is the activation function that measures the difference between the predicted value and the actual value by calculating the mean square error, and the output of the output layer is calculated as: in, is the output of the jth neuron in the hidden layer, is the weight from the hidden layer to the output layer, is the bias of the output layer neurons, M is the number of hidden layer neurons, and the mean square error (MSE) is used as the loss function. The formula is: Where N is the number of samples, y i is the actual value of the ith sample, is the predicted value of the i-th sample. During the training process, the gradient of the loss function with respect to each weight and bias is calculated through the back propagation algorithm, and the gradient descent method is used to update these parameters to minimize the loss function. The performance of the model is evaluated using a data set that did not participate in the training. The trained neural network model is applied to actual scenarios to predict the interference to the power lines during the construction process, and corresponding measures are taken to reduce the interference based on the prediction results.

8. The method for risk assessment of surrounding environment based on covering and excavating subway foundation pit according to claim 1 is characterized in that: The S400 includes: S401, comprehensively evaluate the impact of foundation pit excavation on the safety of surrounding buildings, water supply pipelines, and power facilities, and construct an assessment model based on the assessment objectives and data characteristics; S402, based on data and expert opinions, identify the main risk factors, quantify the risk factors using linear regression, determine their probability of occurrence and impact, analyze the interaction and correlation between factors, calculate the risk level and probability of occurrence of each risk factor, combine all risk factors to obtain the overall risk level, and divide it into three risk levels: low, medium, and high; S403, according to the risk assessment results and actual situation, set reasonable warning thresholds for different risk levels and risk factors, and formulate corresponding warning strategies for each warning threshold; S404: Once the monitoring data exceeds the preset warning threshold, the warning signal is immediately triggered, and notification and response are carried out according to the warning strategy.

9. The method for risk assessment of surrounding environment based on covering and excavating subway foundation pit according to claim 8 is characterized in that: S402 uses linear regression to quantify the risk factors to determine their occurrence probability and impact, and identifies the main risk factors that affect the safety of surrounding buildings, pipelines and power facilities as independent variables X, Y i Represents the value of the dependent variable. The general form of the linear regression model is: Y = β0 + β1X1 + β2X2 + ... + β n X n +∈, where β0 is the intercept, which represents the value of the dependent variable Y when all independent variables are 0, β1,β2,…,β n is the regression coefficient, which indicates the influence of each independent variable on the dependent variable. ∈ is the error term, which represents the part that cannot be explained by the model. The least squares method is used to estimate the model parameters. According to the obtained regression coefficient The absolute value of the regression coefficient reflects the degree of influence of the risk factor on the dependent variable. The established linear regression model is used to predict the probability of occurrence and the degree of influence under new circumstances.

10. The method for risk assessment of surrounding environment based on covering and excavating subway foundation pit according to claim 1, characterized in that: The S500 includes: S501, collect various data from the construction site in real time, clean, integrate and pre-process the collected data, conduct in-depth analysis of the processed data, and identify potential risks and problems; S502: Dynamically adjust the existing risk assessment model based on new data and actual conditions, verify the newly adjusted risk assessment model, and formulate targeted control measures; S503, regularly evaluate the effectiveness of the implementation of control measures, and continuously optimize the monitoring and optimization process based on the evaluation results and feedback.

Images