Method and system for monitoring, early warning and emergency response of manual hole digging pile construction

Abstract

The invention relates to the technical field of data analysis, and discloses a manual hole digging pile construction monitoring early warning and emergency response method and system. The method comprises the steps that pile position parameters and environment data are collected to establish a database; arranging a sensor to obtain monitoring data; identifying risks through a convolutional neural network to generate a prediction matrix; performing quantitative grading to form a four-level early warning signal diagram; formulating a targeted emergency disposal process; response is executed, and processing effects are recorded to form a data recording table. According to the method, the risk mode is recognized in time from the monitoring data through the intelligent analysis processing technology, then the targeted emergency disposal scheme is automatically generated based on the early warning result, a monitoring-early warning-response-evaluation complete closed-loop management system is formed, and the accuracy and effectiveness of manual hole digging pile construction safety management are improved.

Description

Technical Field

[0001] This application relates to the technical field of data analysis, and particularly to a method and system for construction monitoring, early warning and emergency response of manually dug piles. Background Art

[0002] During the traditional construction process of manually dug piles, project safety monitoring and emergency response mainly rely on manual experience judgment and regular inspection methods for management. Construction units usually adopt methods such as manually measuring the inclination of the pile body, periodically checking the geological conditions around the pile position, and adjusting the construction plan according to the weather forecast to monitor construction safety. In terms of emergency handling, standardized emergency plans are mostly adopted, lacking customized treatment plans for the specific engineering characteristics and pile position differences. In the prior art, there are also methods of using a single sensor for local monitoring, such as using an inclinometer to monitor the inclination of the pile body, or using an earth pressure cell to monitor the changes in the soil around the pile. However, these monitoring means are often independent of each other, lacking systematic data integration and analysis processing.

[0003] However, this traditional monitoring, early warning and emergency response method has obvious deficiencies. First, manual experience judgment is highly subjective and difficult to achieve standardization and precision; second, regular inspections have time intervals and cannot achieve continuous monitoring, easily missing key risk signals; third, the single-sensor monitoring method is difficult to comprehensively reflect the complex state changes of the pile body and the surrounding environment; fourth, it lacks the ability to intelligently analyze and process monitoring data and cannot identify risk patterns in a timely manner from a large amount of data; fifth, the standardized emergency plan lacks pertinence and is difficult to formulate accurate treatment plans according to the characteristics and environmental conditions of different pile positions; finally, the emergency response lacks closed-loop management and the evaluation of the treatment effect is insufficient, making it difficult to form an experience accumulation and continuous improvement mechanism. Summary of the Invention

[0004] This application provides a method and system for construction monitoring, early warning and emergency response of manually dug piles, which is used to timely identify risk patterns from monitoring data through intelligent analysis and processing technology, and then automatically generate targeted emergency treatment plans based on the early warning results, forming a complete closed-loop management system of monitoring - early warning - response - evaluation, and improving the accuracy and effectiveness of the construction safety management of manually dug piles.

[0005] In a first aspect, the present application provides a method for construction monitoring, early warning and emergency response of manually dug piles. The method for construction monitoring, early warning and emergency response of manually dug piles includes: establishing a construction environment database of manually dug piles by collecting data such as pile position numbers, pile diameters, pile lengths, excavation sequences, and terrain slopes, relative elevations, and annual rainfall amounts in the project area; according to the construction environment database of manually dug piles, arranging inclinometers, strain sensors, and soil moisture sensors around the pile body to obtain a pile body stability monitoring data set; using the pile body stability monitoring data set, performing pattern recognition on the pile body stress state and soil deformation characteristics through a convolutional neural network to generate a pile position risk prediction matrix; according to the pile position risk prediction matrix, quantitatively grading the risk levels to form a four-level early warning signal distribution map of blue, yellow, orange, and red; based on the four-level early warning signal distribution map, formulating corresponding emergency response procedures for different pile position characteristics and environmental conditions to generate a list of special emergency measures for pile positions; and according to the list of special emergency measures for pile positions, implementing corresponding-level emergency responses, and forming an emergency response data record form by recording key data and measure effects during the disposal process.

[0006] In a second aspect, the present application provides a system for construction monitoring, early warning and emergency response of manually dug piles. The system for construction monitoring, early warning and emergency response of manually dug piles includes: A collection module for establishing a construction environment database of manually dug piles by collecting data such as pile position numbers, pile diameters, pile lengths, excavation sequences, and terrain slopes, relative elevations, and annual rainfall amounts in the project area; An acquisition module for arranging inclinometers, strain sensors, and soil moisture sensors around the pile body according to the construction environment database of manually dug piles to obtain a pile body stability monitoring data set; An identification module for performing pattern recognition on the pile body stress state and soil deformation characteristics through a convolutional neural network using the pile body stability monitoring data set to generate a pile position risk prediction matrix; A quantification module for quantitatively grading the risk levels according to the pile position risk prediction matrix to form a four-level early warning signal distribution map of blue, yellow, orange, and red; A formulation module for formulating corresponding emergency response procedures for different pile position characteristics and environmental conditions based on the four-level early warning signal distribution map to generate a list of special emergency measures for pile positions; An execution module for implementing corresponding-level emergency responses according to the list of special emergency measures for pile positions and forming an emergency response data record form by recording key data and measure effects during the disposal process.

[0007] The third aspect of the present invention provides a computer device, comprising: a memory and at least one processor, wherein instructions are stored in the memory; the at least one processor invokes the instructions in the memory to cause the computer device to execute the above-mentioned method for construction monitoring, early warning and emergency response of manually dug piles.

[0008] The fourth aspect of the present invention provides a computer-readable storage medium, wherein instructions are stored in the computer-readable storage medium, and when it runs on a computer, it causes the computer to execute the above-mentioned method for construction monitoring, early warning and emergency response of manually dug piles.

[0009] In the technical solution provided by this application, by collecting the pile position numbers, pile diameters, pile lengths, excavation sequences, as well as terrain slopes, relative height differences, and annual rainfall data within the project area, an artificial dug pile construction environment database is established, realizing the systematic management and rapid retrieval of key project parameters, and providing comprehensive basic data support for subsequent monitoring and early warning. Inclination measuring instruments, strain sensors, and soil moisture sensors are arranged around the pile body to obtain the pile body stability monitoring data set, solving the limitations of traditional single-parameter monitoring and realizing the all-round and multi-dimensional real-time monitoring of the pile body state and the surrounding environment. Using the pile body stability monitoring data set, through a convolutional neural network, pattern recognition is carried out on the stress state of the pile body and the deformation characteristics of the soil mass to generate a pile position risk prediction matrix, effectively overcoming the subjectivity and uncertainty of traditional manual experience judgment, and significantly improving the accuracy and timeliness of risk identification. As a deep learning algorithm specifically for processing grid-structured data, the convolutional neural network can automatically extract spatio-temporal features and identify complex data patterns. Its application in this solution fully exploits the advantages of the algorithm in processing time-series data and pattern recognition, providing strong technical support for risk prediction. According to the pile position risk prediction matrix, the risk levels are quantitatively classified to form a four-level early warning signal distribution map of blue, yellow, orange, and red, transforming the abstract risk data into an intuitive visual presentation, which is convenient for managers to quickly grasp the overall risk situation of the project. Based on the four-level early warning signal distribution map, corresponding emergency response procedures are formulated according to different pile position characteristics and environmental conditions, generating a list of special emergency measures for pile positions, realizing the precision and personalization of the emergency plan, and greatly enhancing the pertinence and effectiveness of emergency response. According to the list of special emergency measures for pile positions, the corresponding level of emergency response is executed. By recording the key data and measure effects during the disposal process, an emergency response data record form is formed, establishing a closed-loop management mechanism, and providing a data basis for continuous improvement and experience accumulation. Overall, this solution constructs a technical system for artificial dug pile construction monitoring, early warning, and emergency response driven by multi-parameter fusion and intelligent analysis, transforming the traditional experience-based safety management into data-driven precision management, effectively solving key technical problems such as untimely risk identification and non-targeted emergency measures during the project construction process, and significantly improving the construction safety management level and risk control ability. Brief Description of the Drawings

[0010] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for the description of the embodiments will be briefly introduced below. Obviously, the drawings in the following description are some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0011] Figure 1Schematic diagram of an embodiment of the method for construction monitoring, early warning and emergency response of manually dug piles in the embodiments of the present application; Figure 2 Schematic diagram of an embodiment of the system for construction monitoring, early warning and emergency response of manually dug piles in the embodiments of the present application; Figure 3 Schematic block diagram of the structure of a computer device in the embodiments of the present invention. Detailed implementation manners

[0012] The embodiments of the present application provide a method and a system for construction monitoring, early warning and emergency response of manually dug piles. Terms such as "first", "second", "third", "fourth", etc. (if any) in the specification, claims and the above-mentioned drawings of the present application are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged under appropriate circumstances, so that the embodiments described here can be implemented in an order other than those illustrated or described here. In addition, the terms "include" or "have" and any deformation thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device including a series of steps or units does not necessarily have to be limited to those steps or units clearly listed, but may include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0013] For the convenience of understanding, the specific process of the embodiments of the present application will be described below. Please refer to Figure 1 , an embodiment of the method for construction monitoring, early warning and emergency response of manually dug piles in the embodiments of the present application includes: Step S101: Establish a construction environment database for manually dug piles by collecting the pile position number, pile diameter, pile length, excavation sequence, and terrain slope, relative elevation difference, and annual rainfall data in the project area; Step S102: According to the construction environment database for manually dug piles, deploy inclinometers, strain sensors, and soil moisture sensors around the pile body to obtain a pile body stability monitoring data set; Step S103: Use the pile body stability monitoring data set to perform pattern recognition on the pile body stress state and soil deformation characteristics through a convolutional neural network to generate a pile position risk prediction matrix; Step S104: According to the pile position risk prediction matrix, quantitatively classify the risk levels to form a four-level early warning signal distribution map of blue, yellow, orange, and red; Step S105: Based on the four-level early warning signal distribution map, formulate corresponding emergency response procedures for different pile position characteristics and environmental conditions to generate a list of special emergency measures for pile positions; Step S106: According to the special emergency measures list for pile positions, execute the corresponding level of emergency response, and form an emergency disposal data record form by recording the key data and measure effects during the disposal process.

[0014] It can be understood that the execution subject of this application can be a system for construction monitoring, early warning and emergency response of manually dug piles, or a terminal or a server. Specifically, it is not limited here. In this embodiment of the application, the server is taken as the execution subject for illustration.

[0015] Specifically, collect the pile position numbers, pile diameters, pile lengths, excavation sequences, and terrain slopes, relative height differences, and annual rainfall data in the project area, and establish a construction environment database for manually dug piles. Perform geographic information calibration on the bridge pile foundation plan in the project area to extract the pile position spatial coordinates, read the left or right position information, specific pile foundation numbers, accurate pile diameter and pile length values of each pile number from the pile position statistical table. At the same time, collect the natural slope range value of the on-site slope and calculate the relative height difference through the ground natural elevation section data in the bridge area, and then collect the historical hydrological and meteorological data such as the regional annual average rainfall and the maximum annual rainfall. Finally, establish a pile position - terrain - climate correlation data table through hierarchical classification and collation to form a construction environment database for manually dug piles.

[0016] According to the established artificial bored pile construction environment database, inclinometers, strain sensors and soil moisture sensors are arranged around the pile body to obtain the pile body stability monitoring data set. In the specific implementation, the key monitoring point distribution plan is determined according to the pile position number and spatial position information in the database, and then three groups of inclinometers are evenly arranged at an angle of 120° to form a three-dimensional monitoring network for pile body inclination, and strain sensors are pre-buried at key depth positions inside the pile body to construct a pile body force distribution monitoring point matrix. According to the terrain slope and relative height difference data, soil moisture sensors are arranged in the upper soil of the slope to establish a geological stability monitoring network. According to the seasonal law of annual rainfall data analysis, the monitoring frequency is dynamically adjusted to generate an adaptive sampling strategy table. Through the industrial Internet of Things technology, the data of each sensor is collected to the central processing unit and integrated to form a pile body stability monitoring data set. Using the pile body stability monitoring data set, the pile body stress state and soil deformation characteristics are pattern recognized through convolutional neural networks to generate a pile position risk prediction matrix. The process first performs time series segmentation on the monitoring data set to form a data time window sequence, and standardizes the inclination, strain value and soil moisture data through normalization to generate a feature standardized data set. A multi-channel feature map is constructed according to the pile position number and sensor type to form a pile monitoring feature tensor. A three-layer convolution structure is used to extract spatial and temporal features to obtain the pile state feature vector. According to the risk pattern library in the expert knowledge base, the pattern matching algorithm is used to identify the pile stress state and soil deformation characteristics to generate a risk probability distribution. Finally, the risk probability distribution is organized into a matrix form according to the pile position number to construct a pile position risk prediction matrix.

[0017] According to the pile position risk prediction matrix, the risk levels are quantitatively classified to form a distribution map of four - level warning signals: blue, yellow, orange, and red. The specific process includes extracting the risk probability values of each pile position in the pile position risk prediction matrix to establish a pile position risk probability sequence, constructing a four - level risk discrimination threshold table in combination with the engineering safety threshold standard, using this table to classify and judge the pile position risk probability sequence to obtain the pile position risk level classification result, mapping the four - level risk identifiers in the result to the pile numbers to form a corresponding table of pile position risk levels, generating pile position risk spatial distribution data based on this table and the spatial position information in the bridge pile foundation plan, and finally presenting it in a visual way by drawing a four - level warning signal distribution map on the bridge pile foundation plan. Based on the four - level warning signal distribution map, corresponding emergency response procedures are formulated for different pile position characteristics and environmental conditions, and a list of special emergency measures for pile positions is generated. In this step, the warning level information of each pile position is extracted from the warning signal distribution map to construct a pile position warning level mapping table. Combining this table with the pile diameter, pile length, and excavation sequence data in the manual dug pile construction environment database, a pile position characteristic risk correlation matrix is generated. Through the cross - analysis of this matrix with the terrain slope and relative elevation difference data, high - risk geological environment areas are identified to form a geological risk zoning map. According to this map and the annual rainfall data, a differential emergency response procedure framework is established to form a four - level and sixteen - category emergency plan library. From this library, corresponding technical measures, personnel deployment, and resource allocation plans are extracted according to the pile number and warning level to organize and form a preliminary emergency measure combination, and it is optimized and evaluated for feasibility according to the pile position to generate a list of special emergency measures for pile positions.

[0018] According to the list of special emergency measures for pile positions, implement the corresponding level of emergency response. By recording the key data and measure effects during the disposal process, an emergency response data record form is formed. During specific implementation, according to the warning level in the list, start the corresponding level of emergency response instruction to establish an emergency command communication network. Implement targeted disposal on the warned pile positions according to the technical measures part in the list to generate a real - time disposal operation log. Use on - site monitoring equipment to track and record the changes in pile body inclination, strain value, and soil moisture content during the disposal process to form a disposal effect monitoring curve. Compare and analyze it with the data in the pile position risk prediction matrix before disposal to calculate the risk change rate. Evaluate the effectiveness of the disposal measures according to the risk change rate and the disposal operation log to generate a measure effect evaluation result. Integrate all the results to form a structured emergency response data record form.

[0019] In the embodiments of the present application, by collecting the pile position numbers, pile diameters, pile lengths, excavation sequences, and terrain slopes, relative height differences, and annual rainfall data in the project area, an artificial dug pile construction environment database is established, realizing the systematic management and rapid retrieval of key project parameters, and providing comprehensive basic data support for subsequent monitoring and early warning. Inclination measuring instruments, strain sensors, and soil moisture sensors are arranged around the pile body to obtain the pile body stability monitoring data set, solving the limitations of traditional single-parameter monitoring and realizing the all-round and multi-dimensional real-time monitoring of the pile body state and the surrounding environment. Using the pile body stability monitoring data set, a convolutional neural network is used to perform pattern recognition on the stress state of the pile body and the deformation characteristics of the soil body, generating a pile position risk prediction matrix, effectively overcoming the subjectivity and uncertainty of traditional manual experience judgment, and significantly improving the accuracy and timeliness of risk identification. As a deep learning algorithm specifically for processing grid-structured data, the convolutional neural network can automatically extract spatio-temporal features and identify complex data patterns. Its application in this solution fully exploits the advantages of the algorithm in processing time-series data and pattern recognition, providing strong technical support for risk prediction. According to the pile position risk prediction matrix, the risk levels are quantitatively graded to form a four-level early warning signal distribution map of blue, yellow, orange, and red, converting the abstract risk data into an intuitive visual presentation, which is convenient for management personnel to quickly grasp the overall risk situation of the project. Based on the four-level early warning signal distribution map, corresponding emergency response procedures are formulated according to different pile position characteristics and environmental conditions, generating a list of special emergency measures for pile positions, realizing the precision and personalization of the emergency plan, and greatly improving the pertinence and effectiveness of emergency response. According to the list of special emergency measures for pile positions, the corresponding level of emergency response is executed. By recording the key data and measure effects during the disposal process, an emergency response data record form is formed, establishing a closed-loop management mechanism, and providing a data basis for continuous improvement and experience accumulation. Overall, this solution constructs a technical system for artificial dug pile construction monitoring, early warning, and emergency response that integrates multiple parameters and is driven by intelligent analysis, transforming traditional experience-based safety management into data-driven precision management, effectively solving key technical problems such as untimely risk identification and lack of pertinence in emergency measures during the project construction process, and significantly improving the construction safety management level and risk control ability.

[0020] In a specific embodiment, the process of executing step S101 may specifically include the following steps: (1) Perform geographic information calibration on the bridge pile foundation plan of the project area and extract the pile position spatial coordinate data; (2) Read the left or right position information, specific pile foundation number, accurate pile diameter, and pile length values corresponding to each pile number from the pile position statistical table; (3) Collect the natural slope range value of the on-site slope and the ground natural elevation section data of the bridge area, and calculate the relative height difference; (4) Collect historical hydrometeorological data such as the annual average rainfall, maximum annual rainfall, minimum annual rainfall, and maximum daily rainfall in the collection area; (5) Classify and organize the collected pile position parameters and environmental data by layer, and establish a correlation data table of pile position - terrain - climate; (6) Through the correlation data table of pile position - terrain - climate, construct the basic parameter layer in the construction environment database of manually dug piles, and complete the construction environment database of manually dug piles.

[0021] Specifically, conduct geographic information calibration on the bridge pile foundation plan of the project area, extract the spatial coordinate data of the pile positions. By digitally processing the bridge pile foundation plan, establish a unified coordinate system, and convert the pile position on the drawing into the standard spatial coordinates. Geographic information calibration refers to establishing a corresponding relationship between the point positions on the plan and the actual geographical locations. During the specific operation, use geographic information system software. After importing the bridge pile foundation plan, select at least three known control points for registration, and then calculate the accurate spatial coordinates of each pile position through the coordinate transformation algorithm to form a pile position spatial coordinate data table, which includes the number, longitude, latitude, and elevation information of each pile position. Read the left or right position information, specific pile foundation number, accurate pile diameter, and pile length values corresponding to each pile number from the pile position statistical table. The pile position statistical table is an important part of the engineering design document, which records the detailed parameters of all bridge pile foundations. During the data extraction process, for each pile number, sequentially read its position (left or right) in the bridge cross-section, the specific number of the pile foundation (such as 2-2, 2-3, 3-0, etc.), and the key values of the designed pile diameter (such as 2.0m, 2.2m) and pile length (such as 18m, 22m, 24m). These data are the basic information for subsequent risk assessment and early warning, and are directly related to the accuracy of the pile body stability analysis.

[0022] Collect the natural slope range values of the on-site slopes and the data of the natural elevation sections of the ground passed by the bridge area, and calculate the relative height difference. The natural slope of the slope refers to the angle between the mountain or slope surface and the horizontal plane, which is obtained through on-site measurement with measuring instruments. The data of the natural elevation sections of the ground passed by the bridge area reflects the terrain undulation of the area passed by the bridge, and the elevation of key points is obtained through leveling measurement. The calculation method of the relative height difference is to take the elevation difference between the highest point and the lowest point in the bridge area, which characterizes the undulation degree of the terrain. For the construction of manually dug piles, the terrain slope and relative height difference are important factors affecting the construction difficulty and safety risks. Especially for the pile positions distributed at different positions on the slope, the risk assessment criteria need to be adjusted differentially according to the slope and height difference. Collect historical hydrometeorological data such as the annual average rainfall, maximum annual rainfall, minimum annual rainfall, and maximum daily rainfall in the area. These data can be obtained from local meteorological departments or hydrological stations, and usually, the observation records of the past ten years are used for statistical analysis. Hydrometeorological data has a significant impact on the safety of manually dug pile construction. Especially during the rainy season construction, a large amount of rainfall will cause the water content of the soil to increase, affecting the slope stability and the construction safety of the pile body. By analyzing the historical rainfall pattern, the concentrated rainy season (such as May - September) can be determined, and then a targeted monitoring densification strategy can be formulated.

[0023] Classify and organize the collected pile position parameters and environmental data in layers to establish a pile position - terrain - climate correlation data table. Classification and organization in layers is a data processing method that organizes and correlates data of different types and sources according to logical relationships. In specific operations, first, take the pile position parameters (number, position, pile diameter, pile length, excavation sequence) as the basic layer; second, correlate the terrain data (slope, elevation, relative height difference) to each pile position to form a pile position - terrain correlation layer; finally, correlate the regional climate data (rainfall, etc.) with the data of the previous two layers to form a pile position - terrain - climate correlation data table. This multi - dimensional correlation table can intuitively reflect the environmental conditions and potential risk factors of each pile position.

[0024] The last step is to construct the basic parameter layer in the construction environment database of manually dug piles through the pile position - terrain - climate correlation data table, and complete the construction environment database of manually dug piles. The basic parameter layer is the core component of the entire database, containing the static parameters and environmental condition information of all pile positions. Structurally store the pile position - terrain - climate correlation data table through a database management system, establish an indexing and query mechanism, and achieve efficient retrieval and call of data. The construction environment database of manually dug piles is the data basis for subsequent monitoring point layout, risk assessment, and early warning judgment, and its integrity and accuracy directly affect the reliability of the early warning system.

[0025] Taking a highway bridge project as an example, there are 4 pile numbers (2, 3, 11, 13) in the project area, corresponding to 8 specific pile foundations (2-2, 2-3, 3-0, 3-1, 11-2, 11-3, 13-2, 13-3). The spatial coordinates of the pile positions were determined by calibrating the geographic information of the bridge pile foundation plan; from the pile position statistics table, it was read that pile numbers 2 and 11 were located on the right side, pile number 3 was located on the left side, and pile number 13 was located on the right side. The pile diameters were 2.0m and 2.2m, respectively, and the pile lengths were 18m, 24m, and 22m, respectively. On-site measurements showed that the natural slope of the bridge area slope ranged from 10° to 35°, and the slopes of different pile positions were different. The natural elevation of the bridge area through the ground was between 1274 and 1310m, and the relative height difference was calculated to be about 36m. The historical data obtained from the local hydrological department show that the average annual rainfall in the region is 1129.5 mm, the maximum annual rainfall is 1569.3 mm, the minimum annual rainfall is 771.5 mm, and the maximum daily rainfall is 226.2 mm. The rainfall is mainly concentrated from May to September each year, accounting for 80% of the annual precipitation. These data are hierarchically classified and sorted according to the logical relationship of pile position-topography-climate, and an associated data table containing pile position foundation parameters, topographic conditions and climate characteristics is established, and finally a database of the construction environment of artificial bored piles is constructed.

[0026] In a specific embodiment, the process of executing step S102 may specifically include the following steps: (1) Determine the distribution plan of key monitoring points based on the pile position numbers and spatial location information in the bored pile construction environment database; (2) According to the key monitoring point distribution plan, three sets of inclination measuring instruments are evenly arranged at an angle of 120° around each pile to form a three-dimensional monitoring network for pile inclination; (3) Based on the layout of the three-dimensional monitoring network for pile inclination, strain sensors are embedded at key depths inside the pile to construct a monitoring matrix for pile force distribution; (4) Combining the pile force distribution monitoring point matrix and terrain slope data, soil moisture sensors are deployed in the upper soil of the slope to establish a geological stability monitoring network; (5) According to the seasonal patterns of the geological stability monitoring network and annual rainfall data analysis, the monitoring frequency is dynamically adjusted to generate an adaptive sampling strategy table; (6) The data collection is guided by the adaptive sampling strategy table, and the data from each sensor is collected to the central processing unit through the industrial Internet of Things technology to form a pile stability monitoring data set.

[0027] Specifically, according to the pile position number and spatial position information in the manual dug pile construction environment database, determine the distribution scheme of key monitoring points. Extract parameters such as the spatial coordinates, pile diameter, pile length, and excavation sequence of each pile position from the environment database, and consider environmental factors such as terrain slope and relative elevation difference. Through comprehensive analysis, determine the pile positions that need to be key monitored. The distribution scheme of monitoring points includes three elements: the number of monitoring points, their positions, and types. For different pile positions, adopt a differential layout strategy. For example, for pile positions with a larger pile diameter, a longer pile length, or located in a steep slope area, appropriately increase the number of monitoring points; for pile positions with a later excavation sequence, consider the influence of the surrounding already excavated pile positions and adjust the layout of the monitoring points. The distribution scheme of key monitoring points is the layout blueprint of the entire monitoring system, directly determining the comprehensiveness and representativeness of the monitoring data.

[0028] According to the distribution scheme of key monitoring points, evenly arrange three groups of inclinometers around each pile body at an angle of 120° to form a three-dimensional monitoring network for pile body inclination. An inclinometer is a precision instrument for measuring the inclination degree of a structure, capable of continuously recording the inclination changes of the pile body during construction. The 120° angle even arrangement is an optimized arrangement to ensure that the inclination state of the pile body can be comprehensively monitored from three different directions. When installing each group of inclinometers, precise positioning is required, the zero point is calibrated, and it is connected to the data acquisition unit. The data of the three groups of inclinometers are converted through spatial geometric relationships to form a three-dimensional monitoring network, which can accurately capture the inclination changes of the pile body in any direction and achieve full-range monitoring. According to the layout position of the three-dimensional monitoring network for pile body inclination, the next step is to embed strain sensors at key depth positions inside the pile body to construct a monitoring lattice for the force distribution of the pile body. Strain sensors are used to measure the deformation of materials under stress and can reflect the internal stress condition of the pile body. The key depth positions usually include three levels: the pile top, the middle part of the pile body, and the pile bottom. According to the different pile lengths, intermediate measuring points can be appropriately increased. The embedding process needs to be carried out during the pile body construction. The sensors are fixed on the steel reinforcement cage and buried into the pile body along with the concrete pouring. Multiple strain sensors are arranged at different depths and different radial positions to form a three-dimensional lattice structure, which can comprehensively monitor the force distribution of the pile body in the depth and radial directions and provide data support for judging the stress concentration areas inside the pile body.

[0029] Combined with the pile body stress distribution monitoring lattice and terrain slope data, soil moisture sensors are arranged in the upper slope soil to establish a geological stability monitoring network. Soil moisture content is a key factor affecting slope stability. Especially during rainfall, the increase in moisture content will significantly reduce the strength of the soil mass. The layout of the geological stability monitoring network takes into account the slope gradient and soil layer distribution. In areas with steeper slopes and above the pile positions, the arrangement of sensors is appropriately densified. By analyzing the terrain slope data, the layout depth and planar distribution of the soil moisture sensors are determined to form a monitoring network covering key areas, providing real-time data for evaluating slope stability. According to the seasonal patterns analyzed from the geological stability monitoring network and annual rainfall data, the monitoring frequency is dynamically adjusted to generate an adaptive sampling strategy table. The adaptive sampling strategy is a data acquisition scheme designed differentially for monitoring requirements in different periods and different environmental conditions. By analyzing historical annual rainfall data, the rainfall concentration period (such as May - September) and the periods with high rainfall intensity are identified, and the sampling frequency is increased during these high-risk periods; during the dry season or when the meteorological conditions are stable, the sampling frequency is appropriately reduced. The adaptive sampling strategy table is presented in matrix form, with the horizontal axis being the time scale (month or weather condition), the vertical axis being the sensor type, and the matrix elements being the corresponding sampling time intervals. This dynamic adjustment mechanism not only ensures the monitoring density during critical periods but also avoids data redundancy and improves the monitoring efficiency.

[0030] Guided by the adaptive sampling strategy table for data acquisition, the data of each sensor are collected to the central processing unit through industrial Internet of Things technology and integrated to form a pile body stability monitoring data set. Industrial Internet of Things technology refers to the Internet of Things solutions applied in industrial environments, including field sensors, data transmission networks, and central processing systems. Various sensors collect data at the time intervals set by the adaptive sampling strategy table, send them to the on-site data concentrator through wireless transmission or wired network, and then transmit them to the central processing unit through industrial Ethernet. The central processing unit cleans, calibrates, and converts the format of the received raw data, removes outliers and noise, unifies the data format, adds time stamps and location tags, and finally integrates to form a structured pile body stability monitoring data set, laying a data foundation for subsequent pattern recognition and risk prediction.

[0031] For example, in a certain bridge project, by analyzing the construction environment database of manually dug piles, it was determined that pile number 3-1 (pile diameter 2.2m, pile length 24m) is located in the steep slope area on the left side, with a slope of 30°, belonging to a high-risk pile position that requires enhanced monitoring. Accordingly, a distribution plan for key monitoring points was formulated, and three groups of high-precision tilt meters were arranged around this pile position at an angle of 120° to monitor the change in the inclination angle of the pile body in real time. Considering the large pile length, five layers of strain sensors were embedded inside the pile body (respectively at the pile top, 1 / 4, 2 / 4, 3 / 4 of the pile length, and the pile bottom), and 4 radially distributed measuring points were arranged in each layer, forming a 5×4 three-dimensional monitoring matrix. Combining with the slope data of 30°, 8 soil moisture sensors with different depths (0.5m - 3m) were arranged on the upper slope of the pile position to form a geological stability monitoring network covering the key sliding surface. Analyzing the annual rainfall data in this area shows that July to August is the peak rainfall period. Accordingly, an adaptive sampling strategy table was generated: during the peak rainfall period, the tilt meters and strain sensors sample once every 30 minutes, and the soil moisture sensors sample once every 15 minutes; during the regular rainy seasons (May - June, September), the sampling intervals are adjusted to 2 hours and 1 hour respectively; during the dry season (October - April), the sampling intervals are further extended to 4 hours and 2 hours. All sensor data is transmitted to the central processing unit in real time through an industrial wireless sensor network. After data cleaning and integration, a structured monitoring data set containing time, position, parameter values, and status identifiers is formed.

[0032] In a specific embodiment, the process of executing step S103 may specifically include the following steps: (1) Perform time series segmentation on the pile body stability monitoring data set to form a data time window sequence; (2) Standardize the inclination, strain value, and soil moisture data in the data time window sequence through data normalization processing to generate a feature standardization data set; (3) Construct a multi-channel feature map according to the pile position number and sensor type for the feature standardization data set to form a pile body monitoring feature tensor; (4) Adopt a three-layer convolution structure to extract the spatial and time features in the pile body monitoring feature tensor to obtain a pile body state feature vector; (5) According to the pile body state feature vector and the risk pattern library in the expert knowledge base, identify the pile body stress state and soil deformation characteristics through a pattern matching algorithm to generate a risk probability distribution; (6) Organize the risk probability distribution into a matrix form according to the pile position number to construct a pile position risk prediction matrix.

[0033] Specifically, perform time series segmentation on the pile body stability monitoring data set to form a data time window sequence. Time series segmentation is to divide the continuously collected monitoring data into a series of relatively independent but continuous data segments according to a fixed or variable time span. The specific operation is to divide the continuous data of the inclinometer, strain sensor, and soil moisture content sensor collected according to the preset time window length (such as 3 hours, 6 hours, or 12 hours) into multiple windows. For example, divide the continuous 24-hour monitoring data into 4 consecutive time windows with a window length of 6 hours. The selection of the time window should ensure both the time continuity of the data and that the amount of data within each window is sufficient to reflect the change of the pile body state, usually determined according to engineering experience and data characteristics. The purpose of time series segmentation is to facilitate subsequent analysis and processing and capture short-term change trends. Standardize the inclination, strain value, and soil moisture content data in the data time window sequence through data normalization processing to generate a feature standardized data set. Data normalization is the process of converting data with different dimensions and value ranges into a unified standard scale, making the data from different sensors comparable. The specific implementation method is to perform maximum-minimum normalization processing on each type of data in each data time window, mapping the original data values to the interval [0, 1]. For each data point, calculate its relative position in the original data range and convert it into a normalized value. For example, for the inclination data, find the maximum and minimum values within the window, and then map each inclination data according to its proportion between the maximum and minimum values to the interval [0, 1]. In addition, for outliers, use the threshold limit method for processing to avoid the influence of extreme values on the normalization result. The normalization processing enables the sensor data of different types and magnitudes to be analyzed and compared on a unified scale.

[0034] Construct a multi-channel feature map for the feature standardized data set according to the pile position number and sensor type to form a pile body monitoring feature tensor. The multi-channel feature map is a data organization form that organizes data from different sources but is interrelated into a multi-dimensional array structure. The specific implementation is to use the data of different types of sensors (such as inclination, strain value, soil moisture content) at each pile position as independent channels, while maintaining the time dimension and spatial position dimension. For each pile position, construct a three-dimensional data structure: the first dimension is the time window sequence, the second dimension is the spatial distribution position of the sensors, and the third dimension is the sensor type channel. In this way, for multiple pile positions in a bridge project, a four-dimensional pile body monitoring feature tensor is finally formed, and its dimension is [the number of pile positions × the number of time windows × the number of spatial positions × the number of sensor types]. This structured data organization method retains the time continuity, spatial correlation, and diversity of sensor types of the data, providing a suitable input format for subsequent convolutional neural network processing.

[0035] A three - layer convolutional structure is used to extract the spatial and temporal features in the pile monitoring feature tensor, obtaining the pile state feature vector. A convolutional neural network is a deep - learning model particularly suitable for processing data with a grid structure. The three - layer convolutional structure means that the network contains three convolutional layers, and each layer performs feature extraction at different levels. The first - layer convolution mainly extracts local spatio - temporal patterns, using small - sized convolutional kernels (such as 3×3) to slide in the time and space dimensions to capture the data change features in a short time and local area; the second - layer convolution performs higher - level feature extraction on the feature map output by the first layer to capture the change patterns at a medium spatio - temporal scale; the third - layer convolution extracts larger - scale and more abstract spatio - temporal features. The feature map is dimension - reduced through pooling operations to screen out the most significant features. After the three - layer convolution processing, the multi - dimensional feature map is converted into a one - dimensional pile state feature vector through a fully - connected layer. This vector contains the key feature information of the pile's stress state and the deformation of the surrounding soil. According to the pile state feature vector and the risk pattern library in the expert knowledge base, the pile's stress state and soil deformation features are identified through a pattern - matching algorithm, generating a risk probability distribution. The expert knowledge base is a knowledge system constructed from engineering experience and historical cases, and the risk pattern library is a collection of various known risk patterns contained therein. The pattern - matching algorithm is a calculation method for comparing the currently observed features with known patterns and evaluating the similarity. In specific implementation, first, the similarity between the pile state feature vector and each pattern in the risk pattern library is calculated, using measurement methods such as cosine similarity and Euclidean distance. For each risk type, its occurrence probability is calculated based on the similarity, forming a probability distribution vector. At the same time, engineering experience rules are fused to make special judgments on specific risk conditions. For example, when the soil moisture content exceeds a certain threshold, the weight of the landslide risk is increased. The finally generated risk probability distribution is a multi - dimensional vector, and each dimension corresponds to the occurrence probability of a risk type.

[0036] The risk probability distribution is organized into a matrix form according to the pile position numbers to construct a pile - position risk prediction matrix. This step systematically organizes the risk probability distributions of each pile position obtained previously to form an intuitive risk representation structure. The specific implementation is to create a two - dimensional matrix. The horizontal axis is the pile position number, and the vertical axis is different risk types. The matrix element value is the probability of a specific risk occurring at the corresponding pile position. This matrix organization method makes the risk information more structured, facilitating subsequent risk level determination and visualization display. For multiple pile positions in a complex project, the risk prediction matrix provides a global - perspective risk distribution map, helping engineering personnel quickly identify high - risk areas.

[0037] Taking a highway bridge project as an example, the process of processing the monitoring data of pile number 3-1 is as follows: Extract the continuous monitoring data of this pile position for 24 hours from the pile body stability monitoring dataset, and perform time series segmentation with a window length of 6 hours to form 4 time windows. Normalize the inclination data within each window. For example, if the maximum inclination value in window 1 is 0.5° and the minimum value is 0.1°, then the original value of 0.3° is normalized to 0.5 after normalization. Similarly, standardize the strain value and soil moisture content data. Then, organize the standardized data into a multi-channel feature map according to the sensor type. The data of the inclinometer is used as channel 1, the data of the strain sensor is used as channel 2, and the data of the soil moisture content is used as channel 3, maintaining the time window and spatial position dimensions to form a [4×5×3] feature tensor (4 time windows, 5 spatial positions, 3 types of sensors). Input this feature tensor into a three-layer convolutional network. The first layer uses 16 3×3 convolutional kernels to extract local features, the second layer uses 32 3×3 convolutional kernels, and the third layer uses 64 3×3 convolutional kernels. Through pooling and fully connected conversion, a 128-dimensional pile body state feature vector is obtained. Calculate the similarity between this feature vector and each pattern in the risk pattern library to identify the risk distribution in the current state of the pile body. For example, the risk probability of pile body inclination is 75%, the risk probability of stress concentration is 60%, and the risk probability of soil instability is 85%. Finally, organize the risk probability distributions of this pile position and other pile positions in the project into a [8×10] risk prediction matrix (8 pile positions, 10 types of risks) to intuitively display the risk status of each pile position.

[0038] In a specific embodiment, the process of executing step S104 may specifically include the following steps: (1) Extract the risk probability values of each pile position in the pile position risk prediction matrix to establish a pile position risk probability sequence; (2) According to the pile position risk probability sequence, combined with the engineering safety threshold standard, construct a four-level risk discrimination threshold table; (3) Perform hierarchical judgment on the pile position risk probability sequence through the four-level risk discrimination threshold table to obtain the pile position risk level classification result; (4) Map the four-level risk identifications of blue, yellow, orange, and red in the pile position risk level classification result to the pile position numbers to form a pile position risk level correspondence table; (5) Generate pile position risk spatial distribution data according to the pile position risk level correspondence table and the spatial position information in the bridge pile foundation plan; (6) Present the pile position risk spatial distribution data in a visual manner on the bridge pile foundation plan, and draw the distribution maps of the four-level warning signals of blue, yellow, orange, and red.

[0039] Specifically, extract the risk probability values of each pile location in the pile location risk prediction matrix to establish a pile location risk probability sequence. The pile location risk prediction matrix is a two-dimensional data structure. The horizontal axis represents the pile location number, and the vertical axis represents different risk types. The matrix element value is the probability of a specific risk occurring at the corresponding pile location. During the extraction process, for each pile location, calculate the comprehensive risk probability value of all its risk types to form a one-dimensional sequence. The comprehensive risk probability calculation uses the weighted average method, considering the severity of different risk types and the engineering attention level, assigning corresponding weights to obtain a single risk assessment value.

[0040] The process of extracting the pile location risk probability values can be expressed by the following formula: ; where, represents the comprehensive risk probability value of pile location p, represents the total number of risk types, represents the weight coefficient of the k-th type of risk, represents the probability value of the k-th type of risk occurring at pile location p in the pile location risk prediction matrix. The weight coefficient is determined according to the severity of the risk type. For example, the weight of the pile body inclination risk may be higher than that of the slight deformation risk. Through this calculation, each pile location has a corresponding comprehensive risk probability value, and the risk probability values of all pile locations are sorted according to the pile location number to form a pile location risk probability sequence.

[0041] According to the pile location risk probability sequence, combined with the engineering safety threshold standard, construct a four-level risk discrimination threshold table. The engineering safety threshold standard is a grading standard formulated based on engineering experience, industry norms, and safety management requirements, used to distinguish different risk levels. The four-level risk discrimination threshold table divides the risk probability values into four intervals, corresponding to four warning levels: blue (attention), yellow (warning), orange (severe), and red (disaster). The threshold setting takes into account both the general engineering standards and the project characteristics, and is adjusted in combination with factors such as the steepness of the slope, pile length, and pile diameter. For example, for pile locations in steep slope (above 30°) areas, the risk threshold can be appropriately reduced to initiate early warnings; while for pile locations with shorter pile lengths and better geological conditions, the threshold can be slightly increased. The threshold table is presented in matrix form, with the rows being the pile location characteristic classifications (such as pile length, geological conditions, etc.) and the columns being the risk probability thresholds corresponding to the four-level warnings.

[0042] The pile position risk probability sequence is classified through a four-level risk discrimination threshold table to obtain the classification result of the pile position risk level. The classification process is to compare the comprehensive risk probability value of each pile position with the corresponding interval in the threshold table to determine its warning level. The specific operation is to select the corresponding row in the threshold table according to the characteristics of the pile position (such as location, pile length, etc.), and then compare the risk probability value of this pile position with the thresholds of the four warning levels to determine which interval it falls into. For example, if the comprehensive risk probability of a certain pile position is 0.75, and the corresponding threshold intervals are: blue (0 - 0.3), yellow (0.3 - 0.5), orange (0.5 - 0.8), red (0.8 - 1.0), then this pile position is judged to be at the orange warning level. In addition, considering the risk escalation mechanism in special cases, such as the risk value rising continuously for multiple periods or a specific risk type exceeding the standard, even if the comprehensive risk probability does not reach the threshold of a higher level, the warning level may also be raised.

[0043] Map the four-level risk identifications of blue, yellow, orange, and red in the pile position risk level classification result to the pile position numbers to form a corresponding table of pile position risk levels. This step is to establish a clear corresponding relationship between the pile positions and the warning levels, facilitating subsequent risk management and emergency response. The corresponding table of pile position risk levels is a structured data table that contains information such as pile position numbers, risk levels, risk probability values, and main risk types. This table is not only a digital expression of warning information but also the basic data for emergency management. For multiple pile positions in a bridge project, through this table, the risk status and warning levels of each pile position can be intuitively understood for targeted management.

[0044] Based on the pile position risk level correspondence table and the spatial position information in the bridge pile foundation plan, generate the spatial distribution data of pile position risks. The bridge pile foundation plan contains the spatial coordinates and relative position relationships of all pile positions. By associating the risk level information with the spatial positions, risk distribution data with geographical information attributes is formed. The specific implementation is to create a data structure that includes longitude and latitude coordinates, elevation, risk level, and risk description. For each pile position in the plan, add the warning level information extracted from the risk level correspondence table to generate a comprehensive spatial risk dataset. This association gives the risk assessment a spatial dimension and helps to discover risk patterns related to the region. For example, the pile positions in areas with a larger slope may generally have a higher risk level. Present the spatial distribution data of pile position risks in a visual manner on the bridge pile foundation plan, and draw the distribution maps of the four-level warning signals of blue, yellow, orange, and red. Visualization is the process of converting abstract data into intuitive images. In this solution, the risk level is expressed in a color-coding manner. The specific implementation is to, based on the digital bridge pile foundation plan, locate according to the spatial coordinates of the pile positions and mark each pile position with the corresponding warning color (blue, yellow, orange, red). The color coding intuitively reflects the risk level and facilitates managers to quickly identify high-risk areas. In addition, interactive functions can be added, such as mouse hovering to display detailed risk information and clicking to view historical trends, etc., to improve the usability of the warning information.

[0045] Taking a certain expressway bridge project as an example, this project has 8 pile foundations (such as 2-2, 2-3, 3-0, 3-1, etc.). By extracting the data in the pile position risk prediction matrix and applying the weighted average method, calculate the comprehensive risk probability value of each pile position. For example, for the pile position 3-1, there are three main risks: pile body inclination (probability 0.75, weight 0.4), stress concentration (probability 0.6, weight 0.3), and soil mass instability (probability 0.85, weight 0.3). The calculated comprehensive risk probability value is 0.75×0.4 + 0.6×0.3 + 0.85×0.3 = 0.735. According to the project safety standard and the characteristics of this pile position (located on a 30° slope with a pile length of 24m), determine the applicable four-level risk thresholds as: blue (0 - 0.3), yellow (0.3 - 0.5), orange (0.5 - 0.7), red (0.7 - 1.0). By comparison, it is found that 0.735 exceeds the orange upper limit of 0.7. Therefore, this pile position is determined to be at the red warning level. Similarly, conduct grading judgments on other pile positions, map the results to form a pile position risk level correspondence table, and then generate spatial distribution data in combination with the coordinate information in the bridge pile foundation plan. Finally, mark the pile position 3-1 in red on the plan and mark the warning levels of other pile positions with different colors, visually showing the risk distribution status of the entire bridge project and providing clear risk warning information for managers.

[0046] In a specific embodiment, the process of executing step S105 may specifically include the following steps: (1) Extract the warning level information of each pile position from the four-level warning signal distribution map, and construct a pile position warning level mapping table; (2) Combine the pile position warning level mapping table with the pile diameter, pile length, and excavation sequence data in the manual dug pile construction environment database to generate a pile position characteristic risk correlation matrix; (3) Through the cross-analysis of the pile position characteristic risk correlation matrix with the terrain slope and relative elevation difference data, identify high-risk geological environment areas and form a geological risk zoning map; (4) According to the geological risk zoning map and annual rainfall data, formulate a differential emergency response process framework and establish a four-level and sixteen-category emergency plan library; (5) From the four-level and sixteen-category emergency plan library, extract the corresponding technical measures, personnel deployment, and resource allocation plans according to the pile position number and warning level, and organize them into a preliminary emergency measure combination; (6) Through the pile position-targeted optimization and feasibility evaluation of the preliminary emergency measure combination, generate a pile position-specific emergency measure list.

[0047] Specifically, extract the warning level information of each pile position from the four-level warning signal distribution map, and construct a pile position warning level mapping table. This process reads the color coding information from the warning signal distribution map through digital image processing technology to identify the warning level of each pile position. The warning level mapping table is a structured data table that contains the pile position number and the corresponding warning level, and is a bridge for the transformation of warning information into emergency response. During the extraction process, for each pile position marker point, analyze its color attribute, and correspond blue, yellow, orange, and red to the four warning levels respectively, and record them in the mapping table. At the same time, add a timestamp information to record the generation time of the warning status, which is convenient for tracking the warning change trend.

[0048] Combined with the pile position warning level mapping table and the data of pile diameter, pile length, and excavation sequence in the manual dug pile construction environment database, a pile position characteristic risk correlation matrix is generated. The pile position characteristic risk correlation matrix is a two-dimensional data structure that describes the relationship between the physical characteristics of the pile position and the risk level. The rows represent different pile positions, and the columns include information such as pile diameter, pile length, excavation sequence, warning level, and risk correlation score. The generation process first extracts the pile diameter, pile length, and excavation sequence parameters of each pile position from the environment database and associates and matches them with the data in the warning level mapping table according to the pile position number. Then, based on engineering experience and expert knowledge, the risk sensitivity under specific physical characteristics is evaluated. For example, the larger the pile diameter and the longer the pile length, the stronger the risk resistance ability; the later the excavation sequence, the higher the risk of being affected by the surrounding already excavated pile positions. Through these correlation analyses, the characteristic risk correlation score of each pile position is calculated to quantify the influence degree of the physical characteristics of the pile position on the risk situation. Through the cross-analysis of the pile position characteristic risk correlation matrix and the terrain slope and relative elevation difference data, high-risk geological environment areas are identified to form a geological risk zoning map. Cross-analysis refers to the process of combining and comparing two different types of data sets to discover potential correlation patterns. The specific implementation is to perform spatial overlay on the data in the pile position characteristic risk correlation matrix and the terrain slope and relative elevation difference data in the environment database to analyze the correlation between the terrain characteristics and the pile position risk. Identify the concentrated distribution of pile positions with a higher warning level in areas with a steeper slope (such as exceeding 25°) or a larger relative elevation difference, and determine these areas as high-risk geological environment areas. The geological risk zoning map uses a regional division method to divide the project area into different blocks according to the risk level and identify them with different colors or patterns to visually display the spatial distribution characteristics of the risk.

[0049] According to the geological risk zoning atlas and annual rainfall data, a differentiated emergency response process framework is formulated, and an emergency plan library with four levels and sixteen categories is established. The emergency plan library with four levels and sixteen categories is a core component of the construction monitoring, early warning and emergency response method for manual dug piles. According to the four early warning levels of blue (attention), yellow (warning), orange (severe) and red (disaster), combined with risk scenarios such as pile inclination, stress anomaly, soil deformation and hydrological problems, a plan system is formed. Specifically, it includes four types of plans under the blue early warning level: slight pile inclination, local stress anomaly, small deformation of the surrounding soil, and slight water seepage; four types of plans under the yellow early warning level: obvious pile inclination, structural stress concentration, medium deformation of the surrounding soil, and water accumulation in the pile; four types of plans under the orange early warning level: severe pile inclination, reduced bearing capacity, decreased slope stability, and surface water intrusion; four types of plans under the red early warning level: pile instability, pile foundation scouring, landslide and debris flow, and foundation pit collapse. Each plan contains six core elements: pile position description, risk judgment criteria, response level, technical disposal measures, personnel responsibility division, and resource requirement list. Each situation may occur under different levels of early warning states, so it is necessary to formulate differentiated disposal plans. During the formulation process, the characteristics of geological risk zoning are considered, and more stringent response measures are formulated for high-risk areas; at the same time, combined with the analysis of annual rainfall data, an enhanced version of the plan is formulated for the rainy season (such as May-September), and special measures such as drainage and slope reinforcement are added. Each set of plans contains six core elements: target pile position description, risk judgment criteria, emergency response level, technical disposal measures, personnel responsibility division, and resource requirement list.

[0050] From the emergency plan library with four levels and sixteen categories, according to the pile position number and early warning level, the corresponding technical measures, personnel deployment and resource allocation plans are extracted, and an initial emergency measure combination is organized. The extraction process uses a combination of precise matching and fuzzy matching. First, according to the early warning level and main risk type of the pile position, the most similar plan template is precisely matched from the plan library; for special situations, a fuzzy matching algorithm is used to comprehensively consider the content of multiple similar plans and generate a combined plan suitable for the current situation. The initial emergency measure combination includes three main parts: technical measures (such as temporary support for the pile body, reinforcement of the surrounding soil, etc.), personnel deployment (such as the number and division of professional and technical personnel), and resource allocation (such as the type and quantity of equipment and materials). This combination method ensures the comprehensiveness of the emergency response, covering all aspects of the disposal process.

[0051] By conducting targeted optimization and feasibility assessment of the preliminary emergency measure combinations for each pile location, a special emergency measure list for pile locations is generated. Targeted optimization of pile locations is a process of adjusting and refining the preliminary measures according to the specific characteristics and environmental conditions of specific pile locations. For example, for pile locations on slopes, slope protection measures are enhanced; for pile locations with longer pile lengths, deep foundation support plans are strengthened. Feasibility assessment evaluates the implementability of emergency measures from three dimensions: technical implementation difficulty, resource availability, and response timeliness. In the assessment process, a combination of quantitative and qualitative methods is used to score and rank each measure, and the optimal plan combination is selected. The finally generated special emergency measure list for pile locations is a detailed action guidance document, which arranges various disposal measures in chronological order and priority, and clarifies the responsible person, completion time limit, and quality requirements, providing clear guidance for on-site emergency disposal.

[0052] Taking a cross-river bridge project as an example, information was extracted from the four-level warning signal distribution map and it was found that the pile number 3-1 was marked with a red warning level. Based on this, a mapping table including the warning status of 8 pile locations was constructed. Combining the information in the environmental database, the pile diameter of this pile location is 2.2m, the pile length is 24m, the excavation sequence is 8, and it is located on the left side. Integrating these data to form a pile location characteristic risk correlation matrix, it was analyzed that the pile locations with long pile lengths and large diameters have relatively high risks, which is inconsistent with the conventional understanding. Further checking the terrain data, it was found that this pile location is located in the middle and lower part of a 30° slope, with a relative height difference of 15m, and is distributed in the steep slope area on the left side of the bridge site together with other high-risk pile locations, forming an obvious high-risk geological area. Considering that the annual rainfall in this area is mainly concentrated from May to September (accounting for 80% of the whole year), and it is currently the rainy season in July, a comprehensive emergency plan of "red warning - pile inclination - slope instability - heavy rain condition" was accurately matched from the four-level and sixteen-category emergency plan library. This plan includes technical measures such as strengthening the temporary support at the pile top, adding peripheral drainage ditches, installing real-time pile inclination monitors, and laying waterproof geotextiles on the upper slope. Five professional technical personnel are arranged for 24-hour shift monitoring, and resource allocations such as 2 excavators and 3 submersible pumps are equipped. Considering the particularity of this pile location (located on a steep slope and with a long pile length), the plan was optimized specifically: adding the measure of grouting and consolidating the soil around the pile, expanding the layout of drainage facilities to a range of 50m upstream, and at the same time evaluating the feasibility of each measure. Finally, a special emergency measure list for pile locations including 12 specific actions was formed, clarifying the disposal order of "drainage first, support second, and reinforcement third" and the division of responsibilities for each measure.

[0053] In a specific embodiment, the process of executing step S106 may specifically include the following steps: (1) According to the warning level in the special emergency measure list for pile locations, initiate the corresponding level of emergency response instructions and establish an emergency command communication network; (2)Implement targeted disposal on the pre - warned pile positions according to the technical measures section in the special emergency measure list for pile positions, and generate real - time disposal operation logs; (3)Track and record the changes in the pile inclination, strain value, and soil moisture content during the disposal process through on - site monitoring equipment, and form a monitoring curve of the disposal effect; (4)Compare and analyze the monitoring curve of the disposal effect with the data in the risk prediction matrix of the pile position before disposal, and calculate the risk change rate; (5)Conduct an effectiveness assessment of the disposal measures based on the risk change rate and the disposal operation logs, and generate the assessment results of the measure effectiveness; (6)Integrate the assessment results of the measure effectiveness, the disposal operation logs, and the monitoring curve of the disposal effect, and summarize them to form a structured emergency disposal data record form.

[0054] Specifically, according to the warning level in the special emergency measure list for pile positions, initiate the corresponding - level emergency response instructions and establish an emergency command communication network. The emergency response instructions are different response levels divided according to the warning level, including four levels: blue, yellow, orange, and red. Each level corresponds to different response scopes, management levels, and resource mobilization authorities. The specific operation is to read the warning level identifier in the special emergency measure list for pile positions, and compare it with the response level definition in the emergency plan library to determine the level of this emergency start. Subsequently, establish an emergency command communication network, including setting up on - site command points, equipping communication equipment, determining the information transmission path, and establishing a contact list, forming a multi - level communication system covering project management, technical personnel, and construction teams to ensure that emergency instructions can be quickly and accurately conveyed and response actions can be carried out in an orderly manner. Implement targeted disposal on the pre - warned pile positions according to the technical measures section in the special emergency measure list for pile positions, and generate real - time disposal operation logs. The technical measures section is the core content of the emergency list, which details the engineering and technical means to deal with specific risks, such as pile reinforcement, drainage and pressure reduction, soil grouting, etc. Targeted disposal means selectively implementing the most effective technical measures according to the specific situation and risk type of the pile position. During the disposal process, record the implementation time, specific operations, participants, and on - site observation of each measure in a structured manner to form real - time disposal operation logs. This log is not only a real - time record of the disposal work but also an important data source for subsequent evaluation and analysis. It is filled in real - time through an electronic form with a unified format to ensure the integrity and timeliness of the data.

[0055] During the disposal process, a field monitoring device is used to track and record the changes in the inclination of the pile body, strain value, and soil moisture content, forming a monitoring curve of the disposal effect. The field monitoring device includes the inclinometer, strain sensor, and soil moisture sensor deployed in Step 2. During the emergency disposal period, the sampling frequency is increased to achieve intensive monitoring of key parameters. Tracking and recording means continuously collecting monitoring data during the disposal process, tracking the change trend of parameters, and judging the real-time effect of the disposal measures. The monitoring curve of the disposal effect presents the monitoring data of the time series in a graphical manner, intuitively showing the changes in parameters before and after the disposal. In data processing, the moving window average method is used to eliminate the interference of short-term fluctuations, highlight the long-term change trend, and at the same time mark the implementation time points of key disposal measures to establish the corresponding relationship between parameter changes and disposal behaviors. The monitoring curve of the disposal effect is compared and analyzed with the data in the risk prediction matrix of the pile position before the disposal, and the risk change rate is calculated. Comparative analysis is a key link in evaluating the disposal effect. By comparing the changes in key parameters before and after the disposal, the actual effect of the disposal measures is quantified. The specific implementation is to compare the risk probability value corresponding to the pile position in the risk prediction matrix of the pile position before the disposal with the latest risk probability calculated from the current monitoring curve. The calculation of the risk change rate uses the percentage change method, that is, the relative difference between the new and old risk probabilities divided by the original risk probability. A negative value indicates a risk reduction, a positive value indicates a risk increase, and the numerical value reflects the degree of change. At the same time, time series stability analysis is carried out to confirm whether the risk change is a stable trend change or a short-term fluctuation, providing a basis for subsequent evaluation.

[0056] Based on the risk change rate and the disposal operation log, evaluate the effectiveness of the disposal measures and generate the evaluation results of the measure effects. The evaluation of the effectiveness of the disposal measures is a systematic analysis process to judge the effects of the implemented measures, comprehensively considering quantitative data and qualitative information. The quantitative evaluation is mainly based on the risk change rate, and the values are divided into five levels: significant improvement, slight improvement, no obvious change, slight deterioration, and significant deterioration; the qualitative evaluation is based on the on-site observation records and the judgments of professionals in the disposal operation log. The combination of the two evaluations forms a comprehensive evaluation, which individually evaluates each technical measure to judge its effectiveness and incorporates it into the experience database for reference in future similar situations. The evaluation results of the measure effects are a structured evaluation report, containing information such as the effect level, duration, and applicable conditions of each measure. Integrate the evaluation results of the measure effects, the disposal operation log, and the disposal effect monitoring curve, and summarize them into a structured emergency disposal data record form. Structured means organizing data according to a unified format and standard, facilitating storage, retrieval, and analysis. The emergency disposal data record form is a comprehensive data document that integrates the key information of the entire process of emergency response, including five main parts: basic information (pile position number, warning level, response time, etc.), disposal process (operation log summary, key nodes), monitoring data (key parameter change curve), effect evaluation (risk change rate, measure effectiveness), and experience summary (successful experience, problem analysis). The data record form is organized in a hierarchical structure, and the information at different levels has a clear attribution relationship, facilitating subsequent query and analysis. At the same time, as an important part of the case library, it provides reference for the disposal of similar situations.

[0057] Taking a highway bridge project as an example, the pile number 3-1 was identified as a red alert after continuous rainfall. The special emergency measure list for the pile position shows that the pile position is located in the middle and lower part of the slope, with a pile diameter of 2.2m and a pile length of 24m. The main risk is the inclination of the pile body caused by slope instability. According to the red alert level, a level-I emergency response was immediately initiated, an emergency command communication network consisting of the project manager, technical general engineer, and professional engineer was established, and a dual-channel communication mechanism through walkie-talkies and mobile phones was established. Technical disposal was carried out in accordance with the measure list: first, a temporary support steel frame was added on the pile top to resist lateral force, then drainage ditches and anti-seepage geomembranes were arranged on the upper slope to reduce rainwater infiltration, and finally, the soil around the pile body was grouted and reinforced. The entire process lasted for 48 hours, forming a real-time disposal log containing 18 specific operations. Key parameters were tracked in real time through on-site monitoring equipment: the pile inclination decreased from the initial 0.43° to 0.28°, the maximum strain value decreased from 425 microstrains to 310 microstrains, and the soil moisture content of the upper slope decreased from 32% to 24%. These data formed an intuitive monitoring curve of the disposal effect. The monitoring data was substituted into the risk assessment model, and the current risk probability was calculated to be 0.58. Compared with 0.82 before disposal, the risk change rate was -29.3%, indicating a significant reduction in risk. Based on this data and the on-site observation records of technicians in the disposal log, the effectiveness of each measure was evaluated: the temporary support structure had a significant effect and inhibited the continuous inclination of the pile body; the drainage measure had a good effect and effectively reduced the soil moisture content; the grouting reinforcement achieved initial results but needed continuous observation. Finally, all the information was integrated into an emergency disposal data record form, which detailed the whole process from warning trigger to risk control, serving as an important file for project safety management and a reference case for subsequent similar situations.

[0058] The method for construction monitoring, warning, and emergency response of manually dug piles in the embodiments of the present application was described above. Next, the system for construction monitoring, warning, and emergency response of manually dug piles in the embodiments of the present application will be described. Please refer to Figure 2 , an embodiment of the system for construction monitoring, warning, and emergency response of manually dug piles in the embodiments of the present application includes: An acquisition module, configured to establish a construction environment database for manually dug piles by acquiring pile position numbers, pile diameters, pile lengths, excavation sequences, and terrain slope, relative height difference, and annual rainfall data within the project area; An acquisition module, configured to deploy inclinometers, strain sensors, and soil moisture sensors around the pile body according to the construction environment database for manually dug piles to obtain a pile body stability monitoring data set; An identification module, configured to use the pile body stability monitoring data set to perform pattern recognition on the pile body stress state and soil deformation characteristics through a convolutional neural network to generate a pile position risk prediction matrix; A quantification module, which is used to quantify and classify the risk levels according to the pile position risk prediction matrix, and form a distribution map of four-level warning signals of blue, yellow, orange and red; A formulation module, which is used to formulate corresponding emergency response procedures for different pile position characteristics and environmental conditions based on the four-level warning signal distribution map, and generate a list of special emergency measures for pile positions; An execution module, which is used to execute the corresponding level of emergency response according to the list of special emergency measures for pile positions, and form an emergency response data record form by recording the key data and measure effects during the disposal process.

[0059] Through the collaborative cooperation of the above-mentioned various components, by collecting the pile position numbers, pile diameters, pile lengths, excavation sequences, as well as terrain slopes, relative height differences, and annual rainfall data within the engineering area, an artificial dug pile construction environment database is established, realizing the systematic management and rapid retrieval of key project parameters, and providing comprehensive basic data support for subsequent monitoring and early warning. Inclinometers, strain sensors, and soil moisture sensors are arranged around the pile body to obtain the pile body stability monitoring data set, solving the limitations of traditional single-parameter monitoring and realizing the all-round and multi-dimensional real-time monitoring of the pile body state and the surrounding environment. Using the pile body stability monitoring data set, the stress state of the pile body and the deformation characteristics of the soil mass are pattern-recognized through a convolutional neural network to generate a pile position risk prediction matrix, effectively overcoming the subjectivity and uncertainty of traditional manual experience judgment, and significantly improving the accuracy and timeliness of risk identification. As a deep learning algorithm specifically designed for processing grid-structured data, the convolutional neural network can automatically extract spatio-temporal features and identify complex data patterns. Its application in this solution fully exploits the advantages of the algorithm in processing time-series data and pattern recognition, providing powerful technical support for risk prediction. According to the pile position risk prediction matrix, the risk levels are quantitatively graded to form a four-level early warning signal distribution map of blue, yellow, orange, and red, converting the abstract risk data into an intuitive visual presentation, which is convenient for management personnel to quickly grasp the overall risk situation of the project. Based on the four-level early warning signal distribution map, corresponding emergency response procedures are formulated for different pile position characteristics and environmental conditions, generating a list of special emergency measures for pile positions, realizing the precision and personalization of the emergency plan, and greatly enhancing the pertinence and effectiveness of emergency response. According to the list of special emergency measures for pile positions, the corresponding level of emergency response is executed. By recording the key data and measure effects during the disposal process, an emergency response data record form is formed, establishing a closed-loop management mechanism, and providing a data basis for continuous improvement and experience accumulation. Overall, this solution constructs a multi-parameter fusion and intelligent analysis-driven artificial dug pile construction monitoring, early warning, and emergency response technology system, transforming the traditional experience-based safety management into data-driven precision management, effectively solving the key technical problems such as untimely risk identification and non-targeted emergency measures during the project construction process, and significantly improving the construction safety management level and risk control ability.

[0060] Referring to Figure 3 , in the embodiment of the present invention, a computer device is further provided. This computer device can be a server, and its internal structure can be as Figure 3As shown in the figure. The computer device includes a processor, a memory, a display screen, an input device, a network interface, and a database connected through a system bus. Among them, the processor of the computer design is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store the corresponding data in this embodiment. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, the above method is implemented.

[0061] Those skilled in the art can understand that Figure 3 the structure shown in the figure is only a block diagram of some structures related to the solution of the present invention, and does not constitute a limitation on the computer device to which the solution of the present invention is applied.

[0062] An embodiment of the present invention also provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the above method is implemented. It can be understood that the computer-readable storage medium in this embodiment can be a volatile readable storage medium or a non-volatile readable storage medium.

[0063] Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be completed by instructing relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above methods. Among them, any reference to a memory, storage, database, or other medium provided by the present invention and used in the embodiments can include non-volatile and / or volatile memories. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or an external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (SSRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM, etc.

[0064] Those skilled in the art can clearly understand that for the convenience and conciseness of description, the specific working processes of the above-described systems, systems and units can refer to the corresponding processes in the foregoing method embodiments, and will not be elaborated herein.

[0065] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in various embodiments of the present invention. The foregoing storage medium includes: various media that can store program codes such as USB flash drives, mobile hard disks, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical discs.

[0066] As described above, the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements for some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the various embodiments of the present application.

Claims

1. A method for construction monitoring, early warning and emergency response of manually dug piles, characterized in that, The method for monitoring, early warning and emergency response of artificial bored pile construction includes: By collecting data on pile location number, pile diameter, pile length, excavation sequence, terrain slope, relative height difference, and annual rainfall in the project area, a database of bored pile construction environment is established; According to the manual bored pile construction environment database, an inclination meter, a strain sensor and a soil moisture sensor are arranged around the pile body to obtain a pile body stability monitoring data set; Using the pile stability monitoring data set, pattern recognition is performed on the pile stress state and soil deformation characteristics through a convolutional neural network to generate a pile position risk prediction matrix; According to the pile position risk prediction matrix, the risk level is quantitatively graded to form a four-level warning signal distribution map of blue, yellow, orange and red; Based on the four-level warning signal distribution map, formulate corresponding emergency response procedures according to different pile position characteristics and environmental conditions, and generate a list of special emergency measures for pile positions; According to the list of special emergency measures for pile positions, implement the corresponding level of emergency response, and form an emergency disposal data record sheet by recording key data and effects of measures during the disposal process.

2. The method for construction monitoring, early warning and emergency response of manually dug piles according to claim 1, wherein The manual bored pile construction environment database is established by collecting data on pile position number, pile diameter, pile length, excavation sequence, terrain slope, relative height difference, and annual rainfall in the project area, including: Carry out geographic information calibration on the plan view of the bridge pile foundation in the project area and extract the spatial coordinate data of the pile position; Read the left or right position information, specific pile foundation number, precise pile diameter and pile length values ​​corresponding to each pile number from the pile position statistics table; Collect the natural slope range value of the on-site slope and the natural elevation data of the ground through which the bridge passes, and calculate the relative height difference; Collect historical hydrological and meteorological data such as regional average annual rainfall, maximum annual rainfall, minimum annual rainfall and maximum daily rainfall; The collected pile position parameters and environmental data are classified and sorted in layers to establish a pile position-topography-climate correlation data table; Through the pile position-topography-climate correlation data table, the basic parameter layer in the artificial bored pile construction environment database is constructed to complete the artificial bored pile construction environment database.

3. The method for construction monitoring, early warning and emergency response of manually dug piles according to claim 1, characterized in that According to the manual bored pile construction environment database, an inclination meter, a strain sensor and a soil moisture sensor are arranged around the pile body to obtain a pile body stability monitoring data set, including: Determine the key monitoring point distribution plan according to the pile position number and spatial position information in the manual bored pile construction environment database; According to the key monitoring point distribution plan, three groups of inclination measuring instruments are evenly arranged around each pile body at an angle of 120° to form a three-dimensional monitoring network for pile body inclination; According to the layout position of the three-dimensional monitoring network of pile inclination, strain sensors are embedded at key depth positions inside the pile to construct a monitoring matrix of pile force distribution; In combination with the pile force distribution monitoring point matrix and terrain slope data, soil moisture sensors are deployed in the soil mass on the upper part of the slope to establish a geological stability monitoring network; According to the seasonality of the geological stability monitoring network and the annual rainfall data analysis, the monitoring frequency is dynamically adjusted to generate an adaptive sampling strategy table; Guide data acquisition through the adaptive sampling strategy table, and collect sensor data from each sensor to the central processing unit through industrial Internet of Things technology, and integrate and form a pile body stability monitoring data set.

4. The method for construction monitoring, early warning and emergency response of manually excavated piles according to claim 1, characterized in that, Using the pile body stability monitoring data set, perform pattern recognition on the pile body stress state and soil deformation characteristics through a convolutional neural network to generate a pile position risk prediction matrix, including: Perform time series segmentation on the pile body stability monitoring data set to form a data time window sequence; Standardize the inclination, strain value, and soil moisture content data in the data time window sequence through data normalization processing to generate a feature standardized data set; Construct a multi-channel feature map according to the pile position number and sensor type of the feature standardized data set to form a pile body monitoring feature tensor; Adopt a three-layer convolution structure to extract spatial and temporal features in the pile body monitoring feature tensor to obtain a pile body state feature vector; According to the pile body state feature vector and the risk pattern library in the expert knowledge base, identify the pile body stress state and soil deformation characteristics through a pattern matching algorithm to generate a risk probability distribution; Organize the risk probability distribution into a matrix form according to the pile position number to construct a pile position risk prediction matrix.

5. The method for construction monitoring, early warning and emergency response of manually dug piles according to claim 1, characterized in that, According to the pile position risk prediction matrix, quantitatively classify the risk level to form a four-level warning signal distribution map of blue, yellow, orange, and red, including: Extract the risk probability values of each pile position in the pile position risk prediction matrix to establish a pile position risk probability sequence; According to the pile position risk probability sequence, combine with the engineering safety threshold standard to construct a four-level risk discrimination threshold table; Perform a grading judgment on the pile position risk probability sequence through the four-level risk discrimination threshold table to obtain the pile position risk level classification result; Map the four-level risk identifications of blue, yellow, orange, and red in the pile position risk level classification result to the pile position number to form a pile position risk level correspondence table; Generate pile position risk spatial distribution data according to the pile position risk level correspondence table and the spatial position information in the bridge pile foundation plan; Present the pile position risk spatial distribution data in a visual manner on the bridge pile foundation plan, and draw a four-level warning signal distribution map of blue, yellow, orange, and red.

6. The method for construction monitoring, early warning and emergency response of manually dug piles according to claim 1, characterized in that, Based on the four-level warning signal distribution map, formulate corresponding emergency response procedures for different pile position characteristics and environmental conditions to generate a list of pile position specific emergency measures, including: Extract the warning level information of each pile position from the four-level warning signal distribution map to construct a pile position warning level mapping table; Combine the pile position warning level mapping table with the pile diameter, pile length, and excavation sequence data in the manual dug pile construction environment database to generate a pile position characteristic risk correlation matrix; Identify high-risk geological environment areas through cross-analysis of the pile position characteristic risk correlation matrix with terrain slope and relative elevation difference data to form a geological risk zoning map; According to the geological risk zoning map and annual rainfall data, formulate a differential emergency response procedure framework and establish a four-level and sixteen-category emergency plan library; From the four-level and sixteen-category emergency plan library, extract the corresponding technical measures, personnel deployment, and resource allocation plans according to the pile position number and warning level, and organize them to form a preliminary emergency measure combination. Generate a list of special emergency measures for pile positions through pile-position targeted optimization and feasibility assessment of the preliminary emergency measure combination.

7. The method for construction monitoring, early warning and emergency response of manually dug piles according to claim 1, characterized in that, Based on the list of special emergency measures for pile positions, execute the corresponding level of emergency response. By recording the key data and measure effects during the disposal process, form an emergency disposal data record form, including: According to the warning level in the list of special emergency measures for pile positions, initiate the corresponding level of emergency response instruction and establish an emergency command communication network. According to the technical measure part in the list of special emergency measures for pile positions, conduct targeted disposal on the pile positions under warning and generate a real-time disposal operation log. Use on-site monitoring equipment to track and record the changes in pile inclination, strain value, and soil moisture content during the disposal process, and form a disposal effect monitoring curve. Compare and analyze the disposal effect monitoring curve with the data in the pile position risk prediction matrix before disposal, and calculate the risk change rate. Based on the risk change rate and the disposal operation log, conduct an effectiveness assessment of the disposal measures and generate an assessment result of the measure effects. Integrate the assessment result of the measure effects, the disposal operation log, and the disposal effect monitoring curve, and summarize them to form a structured emergency disposal data record form.

8. A system for construction monitoring, early warning and emergency response of manually excavated piles, which is used to implement the method for construction monitoring, early warning and emergency response of manually excavated piles as described in any one of claims 1-7, characterized in that, The system for construction monitoring and early warning and emergency response of manual dug piles includes: A collection module for establishing a construction environment database of manual dug piles by collecting pile position numbers, pile diameters, pile lengths, excavation sequences, and terrain slopes, relative height differences, and annual rainfall data in the project area. An acquisition module for deploying inclinometers, strain sensors, and soil moisture sensors around the pile body according to the construction environment database of manual dug piles to obtain a pile body stability monitoring data set. An identification module for using the pile body stability monitoring data set to perform pattern recognition on the pile body stress state and soil deformation characteristics through a convolutional neural network and generate a pile position risk prediction matrix. A quantification module for quantifying and grading the risk levels according to the pile position risk prediction matrix to form a distribution map of four-level warning signals of blue, yellow, orange, and red. A formulation module for formulating corresponding emergency response procedures for different pile position characteristics and environmental conditions based on the four-level warning signal distribution map and generating a list of special emergency measures for pile positions. An execution module for executing the corresponding level of emergency response according to the list of special emergency measures for pile positions and forming an emergency disposal data record form by recording the key data and measure effects during the disposal process.

9. A computer device, characterized in that, It includes a memory and a processor. The memory stores a computer program that can run on the processor. It is characterized in that when the processor executes the computer program, it implements the method for construction monitoring and early warning and emergency response of manual dug piles according to any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, and when the computer program is run by a processor, the processor is caused to execute the method for construction monitoring, early warning and emergency response of manually dug piles as described in any one of claims 1 to 7.

Images