Method and system for hierarchical early warning and risk control of near-operation underground structure group

Abstract

The application provides a near-operation underground structure group grading early warning and risk control method and system, and the method comprises the following steps: obtaining initial elevation values and real-time elevation values of each monitoring point collected by elevation measurement sensors arranged on each existing structure in a near-operation construction influence area; calculating the settlement amount and the settlement change rate of each monitoring point at the current time according to the initial elevation values and the real-time elevation values, and calculating a combined index according to the settlement amount and the settlement change rate at the current time; calculating an accident possibility score, an exposure frequency score and a consequence severity score at the current time; calculating a comprehensive risk value according to the accident possibility score, the exposure frequency score and the consequence severity score at the current time; determining a risk grade according to the comprehensive risk value, and sending a corresponding risk control strategy to a worker according to the risk grade. The application realizes automatic and objective evaluation and self-adaptive weight distribution of the near-operation construction risk, and can improve the early warning accuracy.

Description

Technical Field

[0001] This invention relates to the field of underground engineering construction technology, and more specifically, to a method and system for graded early warning and risk control of near-operational underground structure groups. Background Technology

[0002] In the field of close-proximity construction in underground engineering, new structures (such as subway stations) often form a close, continuous spatial relationship with existing underground structures such as tunnels and stations. During construction, excavation disturbance can cause stress release in the strata and deformation of existing structures. If not properly controlled, this can lead to structural cracking, excessive settlement, or even operational safety accidents. Therefore, risk assessment and early warning control throughout the entire construction process are crucial. Currently, most risk assessment methods for close-proximity construction rely on human experience and judgment. The quantification of risk factors lacks objective data support, and the weight allocation of each factor remains fixed, making it difficult to dynamically adjust based on the construction stage and historical project data. This results in assessment results that are highly subjective, poorly adaptable, and unable to meet the safety control requirements under complex working conditions. Summary of the Invention

[0003] The purpose of this invention is to provide a method and system for graded early warning and risk control of near-operational underground structure groups, so as to improve the above-mentioned problems.

[0004] To achieve the above objectives, this application provides the following technical solution: On the one hand, embodiments of this application provide a method for graded early warning and risk control of near-operational underground structure groups, the method comprising: Acquire the initial and real-time elevation values ​​of each monitoring point collected by elevation measurement sensors deployed on existing structures within the immediate construction impact zone; The settlement and settlement change rate of each monitoring point at the current moment are calculated based on the initial elevation value and the real-time elevation value. Combined indicators are calculated based on the settlement and settlement change rate at the current moment. The accident probability score at the current moment is calculated based on the combined indicators and the preset risk rule base. The exposure frequency score at the current moment is calculated based on the settlement of all monitoring points on each existing structure. The consequence severity score at the current moment is calculated based on the combined indicators and the preset neural network model. The comprehensive risk value is calculated based on the current accident probability score, exposure frequency score, and consequence severity score; the risk level is determined based on the comprehensive risk value; and the corresponding risk control strategy is retrieved based on the risk level and sent to the staff.

[0005] Secondly, this application provides a graded early warning and risk control system for near-operational underground structure groups, the system comprising: The acquisition module is used to acquire the initial and real-time elevation values ​​of each monitoring point collected by elevation measurement sensors deployed on existing structures in the vicinity of the construction impact zone. The calculation module is used to calculate the settlement and settlement change rate of each monitoring point at the current moment based on the initial elevation value and the real-time elevation value, and to calculate the combined index based on the settlement and settlement change rate at the current moment; to calculate the accident probability score at the current moment based on the combined index and the preset risk rule base; to calculate the exposure frequency score at the current moment based on the settlement of all monitoring points on each existing structure; and to calculate the consequence severity score at the current moment based on the combined index and the preset neural network model. The early warning module is used to calculate a comprehensive risk value based on the current accident probability score, exposure frequency score, and consequence severity score; determine the risk level based on the comprehensive risk value; and retrieve the corresponding risk control strategy based on the risk level and send it to the staff.

[0006] Thirdly, this application provides a graded early warning and risk control device for near-operational underground structure groups, the device comprising a memory and a processor. The memory stores a computer program; the processor executes the computer program to implement the steps of the aforementioned graded early warning and risk control method for near-operational underground structure groups.

[0007] Fourthly, this application provides a readable storage medium storing a computer program, which, when executed by a processor, implements the steps of the above-described method for graded early warning and risk control of near-operational underground structure groups.

[0008] The beneficial effects of this invention are as follows: 1. This invention collects settlement data of existing structures by uniformly deploying static level instruments, and automatically matches the accident probability score with a risk rule base that combines the construction stage name with the settlement change rate threshold. This overcomes the shortcomings of existing technologies that rely on manual experience judgment and lack objective data support for risk factor quantification, and realizes the objectification and automation of risk factor calculation.

[0009] 2. This invention divides historical risk samples into different clusters through cluster analysis and determines the matching cluster to which the current working condition belongs through similarity matching. Based on this, the degree of deviation of the current feature is calculated using standard deviation to obtain the confidence level, and then the historical contribution of each component to the risk level is calculated. The confidence level and historical contribution are then fused and normalized to obtain real-time weights adaptive to the current working condition. Finally, a weighted sum is obtained to obtain the comprehensive risk value. This process is entirely data-driven, requiring no manual setting of fixed weights, and can dynamically adjust the influence of each risk factor according to the construction stage and historical data, improving the accuracy and adaptability of the comprehensive risk value assessment.

[0010] Other features and advantages of the invention will be set forth in the following description, and will be apparent in part from the description, or may be learned by practicing embodiments of the invention. The objects and other advantages of the invention may be realized and obtained by means of the structures particularly pointed out in the written description, claims, and drawings. Attached Figure Description

[0011] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of the present invention and should not be regarded as a limitation on the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0012] Figure 1 This is a schematic diagram of the hierarchical early warning and risk control method for near-operational underground structure groups as described in this embodiment of the invention; Figure 2 This is a schematic diagram of the hierarchical early warning and risk control system for near-operational underground structures as described in this embodiment of the invention; Figure 3 This is a schematic diagram of the hierarchical early warning and risk control equipment for nearby operational underground structures as described in this embodiment of the invention. Detailed Implementation

[0013] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. The components of the embodiments of the present invention described and shown in the accompanying drawings can generally be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to illustrate selected embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without inventive effort are within the scope of protection of the present invention.

[0014] It should be noted that similar reference numerals or letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures. Furthermore, in the description of this invention, terms such as "first," "second," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0015] Example 1 like Figure 1As shown in the figure, this embodiment provides a method for hierarchical early warning and risk control of underground structures in close proximity to operations. The method includes steps S1, S2 and S3.

[0016] Step S1: Obtain the initial and real-time elevation values ​​of each monitoring point collected by the elevation measurement sensors deployed on each existing structure in the vicinity of the construction impact zone. In this step, the immediate construction impact zone refers to the three-dimensional spatial range within which construction disturbances (including stress release in the strata, changes in groundwater seepage, and soil displacement) during the construction of the new underground project can be transmitted to the surrounding existing structures and may cause monitorable deformation or stress changes in the existing structures. In this embodiment, this three-dimensional spatial range is defined as a three-dimensional spatial range extending 30 meters horizontally from the center of the new underground project and vertically from the ground surface to 15 meters below the lowest point of the new underground project. Specifically: Planar range (horizontal direction): Taking the geometric center point of the new underground project as the origin, extend 30 meters longitudinally to both sides, and 30 meters laterally to both sides. Vertical range (elevation direction): Starting from the ground surface, extend downwards to 15 meters below the lowest point of the new underground project.

[0017] Within the aforementioned affected area, identical elevation measurement sensors, specifically hydrostatic levels, were uniformly deployed across all existing structures. After all hydrostatic levels were installed and commissioned, initial elevation values ​​were collected at each monitoring point before construction began (i.e., when the ground was undisturbed by any construction activity). These initial elevation values ​​served as a fixed benchmark for subsequent settlement calculations and remained unchanged throughout the construction monitoring period. Once construction commenced, all hydrostatic levels continuously and automatically measured the elevation values ​​of each monitoring point at the same sampling frequency.

[0018] Through the above method, step S1 achieves unified, continuous, and automated collection of elevation data from all existing structure monitoring points within the immediate construction impact zone.

[0019] Step S2: Calculate the settlement and settlement change rate of each monitoring point at the current moment based on the initial elevation value and the real-time elevation value, and calculate the combined index based on the settlement and settlement change rate at the current moment; calculate the accident probability score at the current moment based on the combined index and the preset risk rule base; calculate the exposure frequency score at the current moment based on the settlement of all monitoring points on each existing structure; calculate the consequence severity score at the current moment based on the combined index and the preset neural network model. In this step, the specific implementation steps for calculating the settlement and settlement change rate of each monitoring point at the current moment based on the initial elevation value and the real-time elevation value, and calculating the combined index based on the settlement and settlement change rate at the current moment include steps S21 and S22. Step S21: Calculate the settlement of each monitoring point at the current moment. This is done by subtracting the initial elevation of each monitoring point from its current elevation, and then multiplying the difference by 1000 to obtain the settlement of each monitoring point at the current moment. Calculate the settlement change rate of each monitoring point at the current moment. This is done by subtracting the settlement of the monitoring point at the previous moment from its current settlement, taking the absolute value of the difference, and then dividing it by the time interval between the two samplings to obtain the settlement change rate of each monitoring point at the current moment. Step S22: Record the maximum value of the settlement at all monitoring points at the current time as the first index; record the maximum value of the settlement change rate at all monitoring points at the current time as the second index; record the difference between the maximum and minimum values ​​of the settlement at all monitoring points at the current time as the third index; combine the first index, the second index and the third index to obtain the combined index at the current time.

[0020] In this step, the first feature F1, the second feature F2, and the third feature F3 are combined to obtain the combined index, which can be understood as: combined index = [F1, F2, F3]; In step S2, the specific implementation steps for calculating the accident probability score at the current moment based on the combined indicators and the preset risk rule base include step S23; Step S23: Obtain the name of the current construction stage and the second feature; match the current construction stage name and the second feature with the preset risk rule base. The risk rule base contains multiple rules. Each rule includes a combination of the construction stage name and the settlement change rate threshold and its corresponding accident probability score. If the current construction stage name is the same as a rule and the second feature is within the settlement change rate threshold corresponding to this rule, the match is considered successful, and the accident probability score corresponding to this rule is used as the accident probability score at the current moment; if the match is unsuccessful, the accident probability score is set to the maximum value among the accident probability scores corresponding to all rules in the risk rule base.

[0021] In this step, the second indicator being within the settlement change rate threshold corresponding to this rule means that the value of the second indicator is greater than or equal to the lower limit of the settlement change rate threshold and less than or equal to the upper limit of the settlement change rate threshold. In this step, for unknown risks, the common practice in engineering safety management is to adopt the worst-case assumption, that is, to assume that the probability of an accident is high. This is because underestimating the risk (e.g., setting it to the minimum) may lead to untimely warnings and inadequate control measures, thus causing a safety accident; while overestimating the risk (setting it to the maximum) will only lead to stricter protective measures without causing safety consequences. Therefore, when an accurate judgment cannot be made, treating it as the maximum risk is a reasonable and safe choice, that is, taking the maximum accident probability score corresponding to all rules in the rule base.

[0022] In step S2, the specific implementation steps for calculating the exposure frequency score at the current moment based on the settlement of all monitoring points on each existing structure include step S24; Step S24: For each existing structure, continuously monitor the settlement of all monitoring points on the existing structure from the start of construction; when the settlement of any monitoring point on the existing structure reaches or exceeds the preset concern threshold for the first time, mark that day as the start date of the disturbance of the existing structure; the number of natural days that have passed from the start date of the disturbance to the current day is recorded as the cumulative number of days; according to the preset mapping relationship between the cumulative number of days and the exposure frequency score, convert the cumulative number of days into the single exposure frequency score of the existing structure; take the maximum value of the exposure frequency score among all existing structures as the exposure frequency score at the current moment.

[0023] In this step, starting from the start of construction, settlement data from all monitoring points on the structure are recorded and checked sequentially in chronological order. When the first monitoring point exceeds the threshold, that day is marked as the start date of the disturbance for the structure. Thereafter, even if other monitoring points on the structure exceed the threshold on later dates, the start date of the disturbance will not be changed.

[0024] The cumulative number of days, calculated from the start date of the disturbance to the current date, is the total number of days including both the start date and the current date. That is, Cumulative Days = Current Date - Start Date of Disturbance + 1; In addition, the preset mapping relationship between the cumulative number of days and the exposure frequency score can be as follows: when the first cumulative number of days is less than or equal to 1 day, the single exposure frequency score is 1; when the first cumulative number of days is greater than 1 day and less than or equal to 3 days, the single exposure frequency score is 2; when the first cumulative number of days is greater than 3 days and less than or equal to 7 days, the single exposure frequency score is 3; when the first cumulative number of days is greater than 7 days and less than or equal to 14 days, the single exposure frequency score is 6; and when the first cumulative number of days is greater than 14 days, the single exposure frequency score is 10.

[0025] In step S2, the specific implementation steps for calculating the severity score of the consequences at the current moment based on the combined indicators and the preset neural network model include step S25; Step S25: Construct a neural network model, which is a feedforward neural network containing one input layer, at least one hidden layer, and one output layer; obtain a training dataset, which consists of multiple historical project samples, each of which includes historical combined indicators and corresponding consequence severity scores; train the neural network model using the training dataset, iteratively updating the network weights using the backpropagation algorithm until the loss function converges, and obtain the trained neural network model; input the combined indicator at the current time into the trained neural network model for forward propagation calculation, and obtain the value output by the output layer as the consequence severity score at the current time.

[0026] In this step, the feedforward neural network consists of one input layer, two hidden layers, and one output layer. The input layer has 3 nodes, the first hidden layer has 8 nodes with ReLU activation, the second hidden layer has 4 nodes with ReLU activation, and the output layer has 1 node with a linear activation function. The loss function is mean squared error (MSE), and the optimizer is Adam.

[0027] Each historical engineering sample includes historical combination indicators and corresponding consequence severity scores. In this step, the consequence severity scores corresponding to the historical combination indicators are given by geotechnical engineering experts. Step S3: Calculate the comprehensive risk value based on the accident probability score, exposure frequency score, and consequence severity score at the current moment; determine the risk level based on the comprehensive risk value; retrieve the corresponding risk control strategy based on the risk level and send it to the staff.

[0028] The specific implementation steps of this step include steps S31-S35; Step S31: Obtain multiple historical risk samples. Each historical risk sample contains a historical feature vector, which consists of three components: accident probability score, exposure frequency score, and consequence severity score. Each historical risk sample contains a corresponding risk level, which includes low risk, moderate risk, high risk, and extremely high risk. Divide the historical risk samples belonging to the same risk level into an initial cluster, and calculate the average value of all historical feature vectors in each initial cluster as the cluster center vector of that initial cluster. Step S32: Calculate the similarity between any two initial cluster center vectors, merge two clusters with a similarity greater than a preset similarity threshold into a new cluster, and calculate the cluster center vector of the new cluster. Repeat this merging process until the similarity between any two cluster centers is no greater than the similarity threshold, and obtain the cluster. Step S33: Combine the accident probability score, exposure frequency score and consequence severity score at the current moment to form the current feature vector. Calculate the similarity between the current feature vector and the cluster center vector of each cluster. Determine the cluster with the highest similarity as the matching cluster and calculate the cluster center vector of the matching cluster. In this step, the average of all historical feature vectors within the matching cluster is calculated and used as the cluster center vector of the matching cluster; In steps S31-S33 above, multiple historical risk samples are first acquired. Historical feature vectors belonging to the same risk level are divided into initial clusters, and then merged by calculating the similarity of the cluster center vectors to form clusters that can represent different risk patterns. Based on this, the three risk scores at the current moment are used to form the current feature vector. By calculating its similarity with the center vectors of each cluster, the historical risk pattern most similar to the current working condition is automatically matched. This process allows the current working condition to be accurately classified without relying on manual judgment, providing a reliable reference benchmark for subsequent weight calculations.

[0029] Step S34: Calculate the standard deviation of each component of all historical feature vectors within the matching cluster; subtract the current feature vector from each component of the cluster center vector of the matching cluster, and then divide by the sum of the standard deviation of that component and the preset minimum deviation constant to obtain the deviation coefficient of each component; add all the deviation coefficients to obtain the sum of the deviation coefficients; calculate the ratio of the deviation coefficient of each component to the sum of the deviation coefficients, and then subtract the ratio from the preset value to obtain the confidence level of each component; calculate the historical contribution of each component; multiply the confidence level of each component by its corresponding historical contribution to obtain the comprehensive influence factor of that component; calculate the sum of the comprehensive influence factors of the three components, and divide the comprehensive influence factor of each component by the sum of the comprehensive influence factors to obtain the real-time weight of that component; In this step, the preset minimum deviation constant is 0.1; the preset value is 1. In this step, the specific implementation steps for calculating the historical contribution of each component include steps S341-S343; Step S341: Record the values ​​of each component of all historical risk samples within the matching cluster as the first sequence, and record the quantified values ​​of the risk level corresponding to all historical risk samples as the second sequence, wherein the risk level is quantified as 1, 2, 3, and 4 according to low risk, general risk, high risk, and extremely high risk respectively. In this step, the first sequence refers to the sequence of all values ​​for the same component (e.g., accident probability score, exposure frequency score, or consequence severity score) across all historical risk samples within the matching cluster. In other words, when calculating the historical contribution of the accident probability score, the first sequence represents the accident probability score of all historical risk samples; when calculating the exposure frequency score, the first sequence represents the exposure frequency score of all historical risk samples; and when calculating the consequence severity score, the first sequence represents the consequence severity score of all historical risk samples.

[0030] The second sequence refers to the sequence composed of the risk level quantification values ​​corresponding to all historical risk samples within the same matching cluster. The risk level quantification rules are: low risk is 1, moderate risk is 2, high risk is 3, and extremely high risk is 4.

[0031] Step S342: Calculate the average of the first series, denoted as the first average; calculate the average of the second series, denoted as the second average; for each historical risk sample within the matching cluster, subtract the first average from the value of the historical risk sample in the first series to obtain the first bias; subtract the second average from the value of the historical risk sample in the second series to obtain the second bias; multiply the first bias and the second bias to obtain the bias product; sum the bias products of all historical risk samples, divide by the number of historical risk samples minus one, and obtain the covariance; Step S343: Calculate the standard deviation of the first series and denot it as the first standard deviation; calculate the standard deviation of the second series and denot it as the second standard deviation; divide the covariance by the product of the first standard deviation and the second standard deviation to obtain the correlation coefficient; take the absolute value of the correlation coefficient to obtain the historical contribution of the component.

[0032] In the above steps, the standard deviation of each component within the matched cluster is calculated, and combined with the difference between the current feature vector and the cluster center vector, the deviation coefficient and confidence level of each component are obtained, thereby quantifying the degree of deviation between the current feature and the cluster center vector. Simultaneously, by using the values ​​of each component within the matched cluster and the quantified risk level as the first and second sequences, respectively, the covariance and standard deviation are calculated to obtain the historical contribution. This process is entirely based on the statistical regularity of historical data, objectively reflecting the actual impact of each component on the final risk level in historical projects, avoiding the arbitrariness of subjective weighting. The confidence level is then multiplied by the historical contribution to obtain the comprehensive influence factor, which is normalized to obtain the real-time weights of the three components. This weight considers both the deviation between the current working condition and the cluster center vector of the matched cluster, and incorporates the actual importance of each component in historical projects, allowing the weights to adaptively adjust with changes in construction stages and monitoring data.

[0033] Step S35: Multiply the current accident probability score, exposure frequency score, and consequence severity score by their real-time weights and sum them to obtain the comprehensive risk value; determine the risk level based on the comprehensive risk value, retrieve the corresponding risk control strategy based on the risk level, and send it to the staff.

[0034] In this step, after obtaining the comprehensive risk value, the system automatically matches the risk level to which the comprehensive risk value belongs according to the preset comprehensive risk value-risk level mapping table (the comprehensive risk value and the risk level correspond one-to-one), and then automatically matches the risk control strategy corresponding to the risk level according to the risk level-risk control strategy mapping table (the risk level and the risk control strategy correspond one-to-one), and sends it to the relevant staff, thereby realizing the whole process management from risk quantification to graded early warning.

[0035] Example 2 like Figure 2 As shown, this embodiment provides a hierarchical early warning and risk control system for near-operational underground structure groups. The system includes an acquisition module 1, a calculation module 2, and an early warning module 3.

[0036] Module 1 is used to acquire the initial and real-time elevation values ​​of each monitoring point collected by elevation measurement sensors deployed on each existing structure in the vicinity of the construction impact zone. Calculation module 2 is used to calculate the settlement and settlement change rate of each monitoring point at the current moment based on the initial elevation value and the real-time elevation value, and to calculate the combined index based on the settlement and settlement change rate at the current moment; to calculate the accident probability score at the current moment based on the combined index and the preset risk rule base; to calculate the exposure frequency score at the current moment based on the settlement of all monitoring points on each existing structure; and to calculate the consequence severity score at the current moment based on the combined index and the preset neural network model. The early warning module 3 is used to calculate the comprehensive risk value based on the accident probability score, exposure frequency score, and consequence severity score at the current moment; determine the risk level based on the comprehensive risk value; and retrieve the corresponding risk control strategy based on the risk level and send it to the staff.

[0037] In one specific embodiment of this disclosure, the computing module 2 further includes a first computing unit 21 and a combination unit 22.

[0038] The first calculation unit 21 is used to calculate the settlement of each monitoring point at the current moment. Specifically, the initial elevation value of each monitoring point is subtracted from the elevation value of the monitoring point at the current moment, and the difference is multiplied by one thousand to obtain the settlement of each monitoring point at the current moment. The settlement change rate of each monitoring point at the current moment is calculated by subtracting the settlement of the monitoring point at the previous moment from the settlement of the monitoring point at the current moment, taking the absolute value of the difference, and then dividing it by the time interval between the two samplings to obtain the settlement change rate of each monitoring point at the current moment. Combination unit 22 is used to record the maximum value of the settlement of all monitoring points at the current time as the first index; record the maximum value of the settlement change rate of all monitoring points at the current time as the second index; record the difference between the maximum and minimum values ​​of the settlement of all monitoring points at the current time as the third index; and combine the first index, the second index and the third index to obtain the combined index at the current time.

[0039] In one specific embodiment of this disclosure, the calculation module 2 further includes a second calculation unit 23.

[0040] The second calculation unit 23 is used to obtain the name of the construction stage and the second feature at the current moment; and to match the current construction stage name and the second indicator with a preset risk rule base. The risk rule base contains multiple rules, each rule including a combination of the construction stage name and the settlement change rate threshold and its corresponding accident probability score. If the current construction stage name is the same as a rule and the second indicator is within the settlement change rate threshold corresponding to this rule, the match is considered successful, and the accident probability score corresponding to this rule is used as the accident probability score at the current moment; if the match is unsuccessful, the accident probability score is set to the maximum value among the accident probability scores corresponding to all rules in the risk rule base.

[0041] In one specific embodiment of this disclosure, the calculation module 2 further includes a third calculation unit 24.

[0042] The third calculation unit 24 is used to continuously monitor the settlement of all monitoring points on each existing structure from the start of construction. When the settlement of any monitoring point on the existing structure reaches or exceeds the preset concern threshold for the first time, the day is marked as the start date of the disturbance of the existing structure. The number of natural days that have passed from the start date of the disturbance to the current day is recorded as the cumulative number of days. According to the preset mapping relationship between the cumulative number of days and the exposure frequency score, the cumulative number of days is converted into the single exposure frequency score of the existing structure. The maximum value of the exposure frequency score among all existing structures is taken as the exposure frequency score at the current moment.

[0043] In one specific embodiment of this disclosure, the calculation module 2 further includes a fourth calculation unit 25.

[0044] The fourth computational unit 25 is used to construct a neural network model, which is a feedforward neural network containing an input layer, at least one hidden layer, and an output layer. It acquires a training dataset consisting of multiple historical project samples, each including historical combined indicators and corresponding consequence severity scores. The neural network model is trained using the training dataset, and the network weights are iteratively updated using the backpropagation algorithm until the loss function converges, resulting in the trained neural network model. The combined indicator at the current time step is input into the trained neural network model for forward propagation calculation, and the output value of the output layer is obtained as the consequence severity score at the current time step.

[0045] It should be noted that the specific methods by which each module performs operations in the system described in the above embodiments have been described in detail in the embodiments related to the method, and will not be elaborated here.

[0046] Example 3 Corresponding to the above method embodiments, this disclosure also provides a graded early warning and risk control device for nearby operational underground structure groups. The graded early warning and risk control device for nearby operational underground structure groups described below can be referred to in correspondence with the graded early warning and risk control method for nearby operational underground structure groups described above.

[0047] Figure 3 This is a block diagram illustrating a graded early warning and risk control device 300 for nearby operational underground structures, according to an exemplary embodiment. Figure 3 As shown, the proximity-operated underground structure group hierarchical early warning and risk control device 300 may include: a processor 301 and a memory 302. The proximity-operated underground structure group hierarchical early warning and risk control device 300 may also include one or more of the following: a multimedia component 303, an I / O interface 304, and a communication component 305.

[0048] The processor 301 controls the overall operation of the proximity-operated underground structure group graded early warning and risk control device 300 to complete all or part of the steps in the aforementioned proximity-operated underground structure group graded early warning and risk control method. The memory 302 stores various types of data to support the operation of the proximity-operated underground structure group graded early warning and risk control device 300. This data may include, for example, instructions for any application or method operating on the proximity-operated underground structure group graded early warning and risk control device 300, as well as application-related data, such as contact data, sent and received messages, images, audio, video, etc. The memory 302 can be implemented using any type of volatile or non-volatile storage device or a combination thereof, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read-Only Memory (EPROM), Programmable Read-Only Memory (PROM), Read-Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk. The multimedia component 303 may include a screen and an audio component. The screen may be, for example, a touchscreen, and the audio component is used to output and / or input audio signals. For example, the audio component may include a microphone for receiving external audio signals. The received audio signals may be further stored in the memory 302 or transmitted via the communication component 305. The audio component also includes at least one speaker for outputting audio signals. I / O interface 304 provides an interface between processor 301 and other interface modules, such as keyboards, mice, and buttons. These buttons can be virtual or physical. Communication component 305 is used for wired or wireless communication between the proximity-operated underground structure group graded early warning and risk control device 300 and other devices. Wireless communication includes, for example, Wi-Fi, Bluetooth, Near Field Communication (NFC), 2G, 3G, or 4G, or a combination thereof. Therefore, the corresponding communication component 305 may include a Wi-Fi module, a Bluetooth module, and an NFC module.

[0049] In an exemplary embodiment, the proximity-operated underground structure group hierarchical early warning and risk control device 300 may be implemented by one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors, or other electronic components to perform the aforementioned proximity-operated underground structure group hierarchical early warning and risk control method.

[0050] In another exemplary embodiment, a computer-readable storage medium including program instructions is also provided. When executed by a processor, these program instructions implement the steps of the aforementioned method for graded early warning and risk control of proximity-operated underground structures. For example, the computer-readable storage medium may be the aforementioned memory 302 including program instructions, which may be executed by the processor 301 of the proximity-operated underground structure group graded early warning and risk control device 300 to complete the aforementioned method for graded early warning and risk control of proximity-operated underground structures.

[0051] Example 4 Corresponding to the above method embodiments, this disclosure also provides a readable storage medium. The readable storage medium described below can be referred to in conjunction with the above-described method for hierarchical early warning and risk control of nearby operational underground structure groups.

[0052] A readable storage medium storing a computer program, which, when executed by a processor, implements the steps of the near-operational underground structure group hierarchical early warning and risk control method described in the above method embodiments.

[0053] Specifically, the readable storage medium can be a USB flash drive, a portable hard drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk, or any other readable storage medium capable of storing program code.

[0054] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A graded early warning and risk control method for underground structure groups in close proximity to operations, characterized in that, include: Acquire the initial and real-time elevation values ​​of each monitoring point collected by elevation measurement sensors deployed on existing structures within the immediate construction impact zone; The settlement and settlement change rate of each monitoring point at the current moment are calculated based on the initial elevation value and the real-time elevation value. Combined indicators are calculated based on the settlement and settlement change rate at the current moment. The accident probability score at the current moment is calculated based on the combined indicators and the preset risk rule base. The exposure frequency score at the current moment is calculated based on the settlement of all monitoring points on each existing structure. The consequence severity score at the current moment is calculated based on the combined indicators and the preset neural network model. The comprehensive risk value is calculated based on the accident probability score, exposure frequency score, and consequence severity score at the current moment; The risk level is determined based on the comprehensive risk value, and the corresponding risk control strategy is retrieved and sent to the staff based on the risk level.

2. The method for graded early warning and risk control of near-operational underground structure groups according to claim 1, characterized in that, Calculate the settlement and settlement change rate of each monitoring point at the current moment based on the initial elevation and real-time elevation values, and calculate the combined index based on the settlement and settlement change rate at the current moment, including: Calculate the settlement at each monitoring point at the current moment by subtracting the current elevation value from the initial elevation value of each monitoring point, and then multiplying the difference by 1000 to obtain the settlement at each monitoring point at the current moment; calculate the settlement change rate at each monitoring point at the current moment by subtracting the settlement at the previous moment from the settlement at the current moment, taking the absolute value of the difference, and then dividing by the time interval between the two samplings to obtain the settlement change rate at each monitoring point at the current moment. The maximum value of the settlement at all monitoring points at the current moment is recorded as the first indicator; the maximum value of the settlement change rate at all monitoring points at the current moment is recorded as the second indicator; the difference between the maximum and minimum values ​​of the settlement at all monitoring points at the current moment is recorded as the third indicator; the first, second, and third indicators are combined to obtain the combined indicator at the current moment.

3. The method for graded early warning and risk control of near-operational underground structure groups according to claim 1, characterized in that, The accident probability score at the current moment is calculated based on a combination of indicators and a preset risk rule base, including: Obtain the name of the current construction stage and its second feature; match the current construction stage name and the second feature with a preset risk rule base. The risk rule base contains multiple rules, each of which includes a combination of the construction stage name and a settlement change rate threshold, along with its corresponding accident probability score. If the current construction stage name is the same as a rule, and the second feature is within the settlement change rate threshold corresponding to this rule, then the match is considered successful, and the accident probability score corresponding to this rule is used as the accident probability score for the current moment; if the match is unsuccessful, then the accident probability score is set to the maximum value among all accident probability scores corresponding to all rules in the risk rule base.

4. The method for graded early warning and risk control of near-operational underground structure groups according to claim 1, characterized in that, The exposure frequency score at the current moment is calculated based on the settlement at all monitoring points on each existing structure, including: For each existing structure, the settlement of all monitoring points on the existing structure is continuously monitored from the start of construction. When the settlement of any monitoring point on the existing structure reaches or exceeds the preset concern threshold for the first time, that day is marked as the start date of the disturbance for the existing structure. The number of natural days that have passed from the start date of the disturbance to the current day is recorded as the cumulative number of days. According to the preset mapping relationship between the cumulative number of days and the exposure frequency score, the cumulative number of days is converted into the single exposure frequency score of the existing structure. The maximum value of the exposure frequency score among all existing structures is taken as the exposure frequency score at the current moment.

5. The method for graded early warning and risk control of near-operational underground structure groups according to claim 1, characterized in that, The severity score of the consequences at the current moment is calculated based on a combination of indicators and a pre-defined neural network model, including: A neural network model is constructed, which is a feedforward neural network containing one input layer, at least one hidden layer, and one output layer. A training dataset is obtained, which consists of multiple historical project samples. Each historical project sample includes historical combined indicators and corresponding consequence severity scores. The neural network model is trained using the training dataset, and the network weights are iteratively updated using the backpropagation algorithm until the loss function converges, resulting in the trained neural network model. The combined indicator at the current time step is input into the trained neural network model for forward propagation calculation, and the value output by the output layer is obtained as the consequence severity score at the current time step.

6. A graded early warning and risk control system for near-operational underground structure groups, characterized in that, include: The acquisition module is used to acquire the initial and real-time elevation values ​​of each monitoring point collected by elevation measurement sensors deployed on existing structures in the vicinity of the construction impact zone. The calculation module is used to calculate the settlement and settlement change rate of each monitoring point at the current moment based on the initial elevation value and the real-time elevation value, and to calculate the combined index based on the settlement and settlement change rate at the current moment; to calculate the accident probability score at the current moment based on the combined index and the preset risk rule base; to calculate the exposure frequency score at the current moment based on the settlement of all monitoring points on each existing structure; and to calculate the consequence severity score at the current moment based on the combined index and the preset neural network model. The early warning module is used to calculate a comprehensive risk value based on the accident probability score, exposure frequency score, and consequence severity score at the current moment. The risk level is determined based on the comprehensive risk value, and the corresponding risk control strategy is retrieved and sent to the staff based on the risk level.

7. The hierarchical early warning and risk control system for near-operational underground structure groups according to claim 6, characterized in that, The calculation module includes: The first calculation unit is used to calculate the settlement of each monitoring point at the current moment. Specifically, the initial elevation value of each monitoring point is subtracted from the elevation value of the monitoring point at the current moment, and the difference is multiplied by one thousand to obtain the settlement of each monitoring point at the current moment. The settlement change rate of each monitoring point at the current moment is also calculated. Specifically, the settlement of the monitoring point at the current moment is subtracted from the settlement of the monitoring point at the previous moment, the absolute value of the difference is taken, and then divided by the time interval between the two samplings to obtain the settlement change rate of each monitoring point at the current moment. The combined unit is used to record the maximum value of the settlement at all monitoring points at the current time as the first index; the maximum value of the settlement change rate at all monitoring points at the current time as the second index; the difference between the maximum and minimum values ​​of the settlement at all monitoring points at the current time as the third index; and to combine the first, second, and third indices to obtain the combined index at the current time.

8. The hierarchical early warning and risk control system for near-operational underground structure groups according to claim 6, characterized in that, The calculation module includes: The second calculation unit is used to obtain the name of the construction stage and the second feature at the current moment; it matches the current construction stage name and the second indicator with a preset risk rule base. The risk rule base contains multiple rules, each rule including a combination of the construction stage name and the settlement change rate threshold and its corresponding accident probability score. If the current construction stage name is the same as a rule and the second indicator is within the settlement change rate threshold corresponding to this rule, the match is considered successful, and the accident probability score corresponding to this rule is used as the accident probability score at the current moment; if the match is unsuccessful, the accident probability score is set to the maximum value among the accident probability scores corresponding to all rules in the risk rule base.

9. The hierarchical early warning and risk control system for near-operational underground structure groups according to claim 6, characterized in that, The calculation module includes: The third calculation unit is used to continuously monitor the settlement of all monitoring points on each existing structure from the start of construction. When the settlement of any monitoring point on the existing structure reaches or exceeds the preset concern threshold for the first time, the day is marked as the start date of the disturbance of the existing structure. The number of natural days that have passed from the start date of the disturbance to the current day is recorded as the cumulative number of days. According to the preset mapping relationship between the cumulative number of days and the exposure frequency score, the cumulative number of days is converted into the single exposure frequency score of the existing structure. The maximum value of the exposure frequency score among all existing structures is taken as the exposure frequency score at the current moment.

10. The hierarchical early warning and risk control system for near-operational underground structure groups according to claim 6, characterized in that, The calculation module includes: The fourth computational unit is used to construct a neural network model, which is a feedforward neural network containing an input layer, at least one hidden layer, and an output layer. It acquires a training dataset consisting of multiple historical project samples, each including historical combined indicators and corresponding consequence severity scores. The neural network model is trained using the training dataset, iteratively updating the network weights using the backpropagation algorithm until the loss function converges, resulting in the trained neural network model. The combined indicator at the current time step is input into the trained neural network model for forward propagation calculation, and the output value of the output layer is obtained as the consequence severity score at the current time step.

Images