Design and construction scheme optimization method and system for subway double-sided expansion and sinking control

Abstract

The present invention discloses a design and construction scheme optimization method and system for heave and sink control of subway double-sided extension. The method collects design parameters, monitoring data and construction management data in the process of subway station extension construction to form a standardized data set. Multi-dimensional fusion analysis is performed on the standardized data set to evaluate the effects of earth excavation, support system and deformation control, and generate target evaluation data. Based on these evaluation results, the optimization algorithm is used to optimize the construction plan of subway double-sided extension to generate optimized sub-plans. The system analyzes the construction progress through the critical path algorithm to form an adjustable schedule, monitors the construction status in real time, dynamically updates the schedule and control measures, and ensures the safety and efficiency of construction. The system can effectively control the deformation problem in the construction process, improve the construction accuracy and management level, and is suitable for improving the quality and safety of subway double-sided extension construction.

Description

Technical Field

[0001] The present application relates to the field of data processing, and in particular to a method and system for optimizing the design and construction scheme for controlling the rise and sink of two-sided extension and construction of subways. Background Art

[0002] With the acceleration of urbanization, the construction scale and difficulty of subway as an important public transportation infrastructure are increasing. The traditional subway station construction progress management method for two-sided subway expansion mainly relies on manual experience and static planning, and uses a single data source for construction monitoring and progress adjustment. This method usually obtains construction site data through regular inspections and sampling inspections, and formulates construction plans through manual analysis. At the same time, the traditional method lacks systematic consideration for the handling of special processes, and often adopts empirical processing methods.

[0003] However, this traditional method has many shortcomings when facing the complex subway station construction environment. First, a single data source is difficult to fully reflect the actual situation of the construction site, which is easy to cause information distortion and decision-making deviation. Secondly, manual analysis is difficult to process massive construction data in a timely manner and cannot quickly respond to emergencies during the construction process. Furthermore, the lack of systematic management of special processes increases construction risks. In addition, static construction plans are difficult to adapt to the dynamically changing construction environment, affecting construction efficiency and quality. Finally, traditional methods are difficult to achieve multi-objective coordinated optimization of construction progress, safety and quality, which restricts the overall benefits of subway station construction. Summary of the invention

[0004] The present application provides a design and construction scheme optimization method and system for subway double-sided extension and construction rise and sink control, which is used to improve the efficiency and accuracy of subway station construction progress management based on data fusion.

[0005] In a first aspect, the present application provides a design and construction scheme optimization method for subway double-sided extension and heave control, the design and construction scheme optimization method for subway double-sided extension and heave control comprises: collecting and preprocessing multi-source data related to subway stations during double-sided extension construction to obtain a standardized data set, wherein the standardized data set comprises design parameters, monitoring data and construction management data;

[0006] Performing multi-dimensional fusion analysis on the standardized data set to obtain target evaluation data, wherein the target evaluation data includes: earth excavation evaluation data, support system evaluation data and deformation control evaluation data;

[0007] Based on the target evaluation data, the preset construction plan is intelligently optimized to obtain an optimized construction plan, wherein the optimized construction plan includes a layered and block excavation construction sub-plan, a counter-pressure island strip drawing construction sub-plan, and an early closure construction sub-plan;

[0008] Based on a preset special process database, the construction progress of the optimized construction plan is analyzed by a critical path algorithm to obtain an adjustable progress plan, wherein the special process database includes: a reserved column process and a reverse loading process;

[0009] Based on the adjustable progress plan, the construction process of the subway station is monitored and analyzed in real time to obtain a real-time state parameter set, wherein the real-time state parameter set includes: surface settlement state, surrounding inclination state and soil stress state;

[0010] Based on the real-time status parameter set, the adjustable progress plan corresponding to the subway station expansion construction process is updated to obtain an updated progress plan and control measures data, wherein the control measures data include active precipitation coupled deformation control measures and station top section dynamic loading control measures.

[0011] In combination with the first aspect, in a first implementation of the first aspect of the present application, the multi-source data related to the subway station during the bilateral extension construction process is collected and preprocessed to obtain a standardized data set, wherein the standardized data set includes design parameters, monitoring data, and construction management data, including:

[0012] Digitally process the design drawings of the subway station to obtain a three-dimensional structural model including reserved columns, prefabricated steel blocks and internal buttress ground-connected walls, and parametrically decompose the three-dimensional structural model to obtain a design parameter set reflecting the structural characteristics of the subway station;

[0013] Real-time monitoring of the subway station construction site is performed through a distributed sensor network to obtain an original monitoring data stream, and noise is eliminated from the original monitoring data stream to obtain a cleaned monitoring data set;

[0014] Performing spatiotemporal alignment processing on the cleaned monitoring data set to obtain multi-dimensional monitoring data in a unified coordinate system, and digitally recording the personnel, equipment and material information at the construction site to obtain original construction management data;

[0015] Structuring the original construction management data to obtain a construction management database containing process dependencies, and unifying the data formats of the design parameter set, multi-dimensional monitoring data and construction management database to obtain a preliminary standardized data set;

[0016] The preliminary standardized data set is cleaned to obtain an intermediate data set from which outliers are removed, and the intermediate data set is completed and normalized to obtain the standardized data set.

[0017] In combination with the first aspect, in a second implementation of the first aspect of the present application, the standardized data set is subjected to a multi-dimensional fusion analysis to obtain target evaluation data, wherein the target evaluation data includes: earthwork excavation evaluation data, support system evaluation data, and deformation control evaluation data, including:

[0018] Performing feature extraction on the design parameters in the standardized data set to obtain a subway station structural feature vector, and performing spatiotemporal correlation analysis on the monitoring data to obtain a multi-dimensional monitoring data association matrix;

[0019] Decomposing the construction management data into process steps to obtain a sequence of key nodes in the construction process, and parametrically modeling the earth excavation scheme based on the structural characteristic vector of the subway station to obtain an earth excavation numerical model;

[0020] Performing stress field simulation on the earth excavation numerical model to obtain earth excavation evaluation data, and dynamically evaluating the stress state of the support system based on the multi-dimensional monitoring data association matrix to obtain support system evaluation data;

[0021] Combined with the sequence of key nodes in the construction process, trend analysis is performed on the deformation monitoring data to obtain a deformation development prediction model, and risk level classification is performed on the deformation development prediction model to obtain deformation control assessment data;

[0022] A multi-criteria decision analysis is performed on the earth excavation assessment data, the support system assessment data and the deformation control assessment data to obtain comprehensive target assessment indicators, and based on the comprehensive target assessment indicators, a subway station construction progress-safety-quality multi-target evaluation system is constructed to obtain the target assessment data.

[0023] In combination with the first aspect, in a third implementation of the first aspect of the present application, based on the target evaluation data, the preset construction plan is intelligently optimized to obtain an optimized construction plan, wherein the optimized construction plan includes a layered and block excavation construction sub-plan, a counter-pressure island strip drawing construction sub-plan, and an early closure construction sub-plan, including:

[0024] Decomposing the target evaluation data into multiple targets to obtain weight coefficients of earth excavation target, support system target and deformation control target, and based on the weight coefficients, preliminarily scoring the preset construction scheme to obtain an initial scoring matrix;

[0025] The initial scoring matrix is ​​optimized and iterated by a genetic algorithm to obtain a set of candidate solutions that meet multi-objective requirements, and the layered and block excavation solutions in the candidate solution set are parameterized to obtain a layered and block model that reflects the excavation depth, block size and excavation sequence;

[0026] Based on the layered and block model, a geological stress field simulation is performed to obtain stress distribution data during the layered and block excavation process, and the layered and block excavation scheme is optimized by the stress distribution data to obtain the layered and block excavation construction sub-scheme;

[0027] Performing geometric optimization on the back-pressure island stripping schemes in the candidate scheme set to obtain the optimal back-pressure island shape and stripping width parameters, and performing stability analysis based on the optimal back-pressure island shape and stripping width parameters to obtain a safety factor curve for the back-pressure island stripping construction;

[0028] The anti-pressure island stripping scheme is optimized by using the safety factor curve to obtain the anti-pressure island stripping construction sub-scheme;

[0029] Optimizing the process of the early closure scheme in the candidate set to obtain the optimal closure time and closure sequence, and performing deformation prediction analysis based on the optimal closure time and closure sequence to obtain deformation control effect evaluation data of the early closure construction;

[0030] Optimizing the early closure scheme by using the deformation control effect evaluation data to obtain the early closure construction sub-scheme;

[0031] The optimized construction plan is obtained by performing data fusion on the layered and block excavation construction sub-plan, the counter-pressure island strip drawing construction sub-plan and the early closure construction sub-plan.

[0032] In combination with the first aspect, in a fourth implementation of the first aspect of the present application, the construction progress analysis of the optimized construction plan is performed based on a preset special process database through a critical path algorithm to obtain an adjustable progress plan, wherein the special process database includes: a reserved column process and a reverse loading process, including:

[0033] Parameterizing the reserved column process in the special process database to obtain a key parameter set for reserved column construction, and based on the key parameter set, performing process decomposition on the reserved column construction to obtain a detailed process network for reserved column construction;

[0034] Performing a time sequence analysis on the reverse loading process in the special process database to obtain the optimal implementation time of the reverse loading, and optimizing the resource allocation of the reverse loading process based on the optimal implementation time to obtain a resource demand curve of the reverse loading;

[0035] Based on the resource demand curve of the reverse loading, a process analysis is performed on the water injection process of the prefabricated steel blocks in the reverse loading process to obtain the quality control points of the water injection of the prefabricated steel blocks, and based on the quality control points, a time estimation is performed on the water injection process of the prefabricated steel blocks to obtain the duration distribution of the water injection process;

[0036] The process of the layered and block excavation construction sub-scheme, the counter-pressure island strip drawing construction sub-scheme and the early closure construction sub-scheme in the optimized construction scheme are integrated to obtain a comprehensive construction network diagram;

[0037] The comprehensive construction network diagram is calculated using a critical path algorithm to obtain the adjustable schedule.

[0038] In combination with the first aspect, in a fifth implementation of the first aspect of the present application, based on the adjustable progress plan, the construction process of the subway station is monitored and analyzed in real time to obtain a real-time state parameter set, wherein the real-time state parameter set includes: surface settlement state, surrounding inclination state and soil stress state, including:

[0039] The adjustable schedule is divided into time windows to obtain a key monitoring time node sequence of the construction process, and based on the key monitoring time node sequence, the data collection frequency of the sensor network at the subway station construction site is optimized to obtain a dynamic sampling scheme;

[0040] The collected original monitoring data is subjected to wavelet transform noise reduction processing to obtain cleaned surface settlement data, cleaned surrounding inclination data and cleaned soil stress data;

[0041] Performing spatiotemporal interpolation on the cleaned surface subsidence data to obtain a continuous surface subsidence field;

[0042] Based on the surface settlement field, the abnormal settlement area is identified by using a curvature analysis algorithm to obtain a surface settlement state assessment result;

[0043] Performing trend analysis on the cleaned surrounding inclination data to obtain a building inclination rate curve, and based on the building inclination rate curve, determining the stability level of surrounding buildings by a threshold discrimination method to obtain a surrounding inclination state assessment result;

[0044] Performing stress path analysis on the cleaned soil stress data to obtain the stress evolution process of key parts, and based on the stress evolution process, calculating the soil stability index through the plastic zone expansion prediction model to obtain the soil stress state assessment result;

[0045] The surface settlement state assessment results, the surrounding inclination state assessment results and the soil stress state assessment results are fused to obtain a real-time state parameter set.

[0046] In combination with the first aspect, in a sixth implementation of the first aspect of the present application, based on the real-time status parameter set, the adjustable schedule corresponding to the subway station expansion construction process is updated to obtain an updated schedule and control measure data, wherein the control measure data includes active precipitation coupled deformation control measures and station top interval dynamic loading control measures, including:

[0047] Performing a risk level assessment on the real-time status parameter set to obtain a current risk index for the subway station construction, and performing a deviation analysis on the adjustable schedule based on the current risk index to obtain a schedule deviation matrix;

[0048] Performing decision analysis on the schedule deviation matrix to obtain a schedule adjustment strategy set, and optimizing time parameters of the adjustable schedule based on the schedule adjustment strategy set to obtain a preliminary updated schedule;

[0049] Re-allocating resources for the preliminary update schedule to obtain an update schedule after resource balancing, and optimizing parameters of the active dewatering system based on the update schedule to obtain a dewatering well layout plan and a pumping rate curve;

[0050] A coupling analysis is performed on the dewatering well arrangement scheme and the pumping rate curve to obtain active dewatering coupled deformation control measures, and based on the update schedule, a loading simulation is performed on the station top interval to obtain a dynamic loading force distribution diagram;

[0051] The dynamic loading force distribution diagram is time-series optimized to obtain the station top interval dynamic loading control measures, and the active precipitation coupled deformation control measures and the station top interval dynamic loading control measures are integrated into a control measure data package.

[0052] In the second aspect, the present application provides a design and construction scheme optimization system for subway double-sided extension and construction heave and sink control, and the design and construction scheme optimization system for subway double-sided extension and construction heave and sink control includes:

[0053] A processing module is used to collect and preprocess multi-source data related to the subway station during the bilateral expansion construction process to obtain a standardized data set, wherein the standardized data set includes design parameters, monitoring data, and construction management data;

[0054] A fusion module is used to perform multi-dimensional fusion analysis on the standardized data set to obtain target evaluation data, wherein the target evaluation data includes: earth excavation evaluation data, support system evaluation data and deformation control evaluation data;

[0055] An optimization module is used to perform intelligent optimization processing on the preset construction plan based on the target evaluation data to obtain an optimized construction plan, wherein the optimized construction plan includes a layered and block excavation construction sub-plan, a counter-pressure island strip drawing construction sub-plan, and an early closure construction sub-plan;

[0056] An analysis module is used to analyze the construction progress of the optimized construction plan based on a preset special process database through a critical path algorithm to obtain an adjustable progress plan, wherein the special process database includes: a reserved column process and a reverse loading process;

[0057] A monitoring module, for performing real-time monitoring and analysis on the construction process of the subway station based on the adjustable progress plan, and obtaining a real-time state parameter set, wherein the real-time state parameter set includes: surface settlement state, surrounding inclination state, and soil stress state;

[0058] An updating module is used to update the adjustable progress plan corresponding to the subway station expansion construction process based on the real-time status parameter set, and obtain an updated progress plan and control measures data, wherein the control measures data include active precipitation coupled deformation control measures and station top interval dynamic loading control measures.

[0059] In the technical solution provided by this application, by collecting and preprocessing the multi-source data related to the subway station during the bilateral expansion construction process, a standardized data set including design parameters, monitoring data and construction management data is obtained, which provides a reliable basis for subsequent analysis. These data are subjected to multi-dimensional fusion analysis to obtain target evaluation data including earthwork excavation evaluation data, support system evaluation data and deformation control evaluation data, realizing comprehensive consideration of key factors. Based on these evaluation data, the preset construction plan is intelligently optimized to obtain an optimized construction plan including layered and block excavation, back pressure island strip extraction and early closure construction sub-plans, which improves construction efficiency and reduces risks. The progress of the optimization plan is analyzed by the critical path algorithm, and special processes such as reserved columns, reverse loading and water injection of steel prefabricated blocks are considered at the same time to obtain an adjustable progress plan, which improves the accuracy and executability of the plan. Based on this plan, the construction process is monitored and analyzed in real time, and a real-time state parameter set including surface settlement state, surrounding inclination state and soil stress state is obtained, which provides a basis for timely discovery of problems. Finally, the progress plan was dynamically updated based on these real-time data, and a control plan was formulated, including active precipitation coupled deformation control and station top interval dynamic loading control measures, to achieve flexible adjustment of the construction plan. This method achieves comprehensive monitoring, accurate evaluation and dynamic optimization of the construction process, significantly improves the efficiency and accuracy of construction management, reduces risks, improves quality, and provides strong support for the safe and efficient construction of subway stations. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0061] Figure 1 A schematic diagram of an embodiment of a method for optimizing the design and construction scheme for controlling the uplift and sinking of subway double-sided extension in an embodiment of the present application;

[0062] Figure 2 This is a schematic diagram of an embodiment of a design and construction scheme optimization system for subway double-sided expansion and subsidence control in an embodiment of the present application. DETAILED DESCRIPTION

[0063] The embodiments of the present application provide a design and construction scheme optimization method and system for subway double-sided expansion and construction heave control. The terms "first", "second", "third", "fourth", etc. (if any) in the specification and claims of this application and the above-mentioned drawings 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 interchangeable where appropriate, so that the embodiments described here can be implemented in an order other than that illustrated or described here. In addition, the terms "including" or "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0064] For ease of understanding, the specific process of the embodiment of the present application is described below. Figure 1 , an embodiment of the design and construction scheme optimization method for subway double-sided expansion and construction heave control in the embodiment of the present application includes:

[0065] Step S101, collecting and preprocessing multi-source data related to the subway station during the bilateral extension construction process to obtain a standardized data set, wherein the standardized data set includes design parameters, monitoring data, and construction management data;

[0066] It is understandable that the execution subject of the present application can be a subway station construction progress management system based on data fusion, or a terminal or a server, which is not limited here. The present application embodiment is described by taking the server as the execution subject as an example.

[0067] Specifically, determine the scope and method of data collection. For design parameters, relevant data such as structural dimensions, material properties, load information, etc. can be extracted through the Building Information Model (BIM) system. These data are usually stored in a structured form in the BIM database and can be obtained through the API interface or data export function. At the same time, the digital processing of unstructured data such as geological survey reports and design drawings is also considered, which involves the application of optical character recognition (OCR) technology and natural language processing (NLP) technology.

[0068] The collection of monitoring data requires the deployment of a variety of sensor networks, including but not limited to strain gauges, inclinometers, displacement sensors, pressure sensors, etc. These sensors transmit data to the central data processing system in real time through wired or wireless networks. For example, distributed fiber optic sensing technology is used to monitor the deformation of subway tunnels, or 3D laser scanning technology is used to obtain real-time geometric information of construction sites. For groundwater level monitoring, automated water level meters can be installed to regularly record changes in groundwater levels. These monitoring data usually exist in the form of time series, and preprocessing operations such as denoising, outlier detection, and interpolation are required.

[0069] The collection of construction management data involves multiple aspects such as project management system, personnel positioning system, material management system, etc. For example, RFID technology is used to track the flow and use of materials, or mobile applications are used to record construction progress and quality inspection results. The video surveillance system at the construction site can also provide a lot of valuable information. Through computer vision technology, data such as the movement trajectory of construction machinery and the activity pattern of workers can be extracted from the video stream. In addition, text data such as construction logs, safety inspection records, and quality control reports also need to be digitized and structured.

[0070] After data collection is completed, the next step is to preprocess and standardize these multi-source heterogeneous data. First, the data from different sources are synchronized in time and aligned in space. For example, use GPS or Beidou positioning system to add unified spatiotemporal tags to all data. For monitoring data, noise reduction processing may be required, such as using wavelet transform or Kalman filtering to remove high-frequency noise in the signal. For data with missing values, interpolation technology or machine learning methods are used to complete the data.

[0071] Data standardization is a key step to ensure the comparability of different types of data. For numerical data, use methods such as min-max standardization or Z-score standardization to map the data to a uniform scale. For categorical data, use methods such as one-hot encoding or label encoding for conversion. Text data can be converted to numerical vectors through word embedding technology. In this process, issues such as data unit unification and format conversion are also considered.

[0072] In order to construct the final standardized data set, a unified data model needs to be designed to integrate design parameters, monitoring data, and construction management data. This data model should be able to reflect the association between different types of data while maintaining sufficient flexibility to accommodate possible future data expansion. For example, an object-oriented data modeling approach is used to treat the various components of a subway station (such as tunnels, stations, ventilation systems, etc.) as objects, each of which contains relevant design parameters, monitoring data, and management information.

[0073] Finally, consider data quality control and data security issues. An automated data quality check process can be set up to identify and process outliers, duplicate data, etc. At the same time, considering the sensitivity of construction data, strict data encryption and access control measures need to be implemented to ensure data security and integrity.

[0074] Step S102: performing multi-dimensional fusion analysis on the standardized data set to obtain target evaluation data, wherein the target evaluation data includes: earth excavation evaluation data, support system evaluation data and deformation control evaluation data;

[0075] Specifically, correlation analysis is performed on various types of data in the standardized data set to identify potential associations between different data. This can be achieved by using statistical methods such as the Pearson correlation coefficient or the Spearman rank correlation coefficient to reveal, for example, the relationship between earthwork excavation progress and surface settlement, or the connection between the installation of the support system and the inclination of surrounding buildings. Dimensionality reduction techniques such as principal component analysis (PCA) or factor analysis are used to extract key features from high-dimensional data. These features may include key influencing factors of earthwork excavation, core performance indicators of the support system, and key parameters of deformation control. Based on these extracted features, multiple special assessment models can be constructed. For example, for earthwork excavation assessment, multivariate regression analysis or support vector machine (SVM) is used to establish a quantitative relationship between excavation depth, excavation area and ground stability. For support system assessment, finite element analysis is combined with measured data to construct a numerical model that can predict the stress condition of the support system. For deformation control assessment, time series analysis methods such as the autoregressive integrated moving average (ARIMA) model are used to predict the development trend of surface settlement and displacement of surrounding buildings.

[0076] In order to integrate these special assessment results into a comprehensive target assessment data set, data fusion technology is used. For example, the Dempster-Shafer evidence theory or fuzzy comprehensive evaluation method is used to integrate the assessment results from different sources. This method can effectively deal with the uncertainty and ambiguity in the assessment process. In addition, machine learning algorithms such as random forests or gradient boosting trees can also be considered to build an integrated model that can simultaneously consider the three aspects of earthwork excavation, support system, and deformation control, and output a comprehensive risk assessment indicator.

[0077] In practical applications, data imbalance may occur, for example, deformation control data may be richer than earthwork excavation data. In response to this situation, oversampling or undersampling techniques are used to balance the data set, or algorithms that can handle imbalanced data, such as weighted SVM or CostSensitive Learning, are used. In addition, considering that subway station construction is a dynamic process, online learning algorithms are also introduced to enable the evaluation model to self-update and optimize as new data is continuously input.

[0078] Step S103: Based on the target evaluation data, the preset construction plan is intelligently optimized to obtain an optimized construction plan, wherein the optimized construction plan includes a layered and block excavation construction sub-plan, a counter-pressure island strip drawing construction sub-plan, and an early closure construction sub-plan;

[0079] Specifically, a multi-objective optimization algorithm, such as the non-dominated sorting genetic algorithm II (NSGA-II) or the multi-objective particle swarm optimization algorithm (MOPSO), is used to take the various indicators in the target evaluation data as optimization targets, while considering multiple aspects such as construction safety, efficiency and cost. For the layered and block excavation construction sub-scheme, the excavation depth, block size and excavation sequence can be used as decision variables, and the optimization algorithm is used to find the best combination that can minimize the surface settlement and the impact of surrounding buildings. In this process, it may be necessary to combine finite element analysis to simulate the impact of different excavation schemes on the stratum stress distribution, so as to consider the constraints of geological conditions in the optimization process. For the back pressure island stripping construction sub-scheme, the shape, size of the back pressure island, and the width and spacing of the stripping can be used as optimization parameters, and the optimal configuration that can minimize the deformation of the foundation pit can be found through numerical simulation and optimization algorithms. This involves fluid-solid coupling analysis to consider the impact of groundwater flow on the stability of the foundation pit. In the optimization of the early closure construction sub-scheme, it is necessary to weigh the impact of the closure timing on the construction progress and deformation control, and use a dynamic programming algorithm to determine the optimal closure sequence and time point, while considering the dependencies of the construction processes and resource constraints. In order to deal with the uncertainty in the optimization process, robust optimization techniques such as scenario analysis or fuzzy optimization methods are introduced to ensure that the optimized construction plan can maintain good performance under various possible construction conditions. In addition, considering the dynamic characteristics of the construction environment, online learning and adaptive optimization algorithms are used to enable the optimized construction plan to be dynamically adjusted according to real-time monitoring data. Finally, by integrating and coordinating these sub-plans, a comprehensive optimized construction plan is obtained, which can not only meet the requirements of various technical indicators, but also adapt to complex construction environments and uncertain factors, thereby improving the safety, efficiency and quality of subway station construction.

[0080] Step S104: Based on the preset special process database, the construction progress of the optimized construction plan is analyzed by the critical path algorithm to obtain an adjustable progress plan, wherein the special process database includes: a reserved column process and a reverse loading process;

[0081] Specifically, a detailed database containing special processes such as the reserved column process and the reverse loading process is established. The database should contain information such as the duration, resource requirements, preconditions and technical requirements of each special process. For example, for the reserved column process, it is necessary to record parameters such as the size of the reserved column, concrete strength grade, and curing time; for the reverse loading process, it is necessary to record information such as the type of loading material, loading height, and loading rate; and for the water injection process of the prefabricated steel blocks, it is necessary to record key parameters such as the water injection pressure, water injection rate, and water injection time. Next, these special processes need to be integrated with the conventional processes in the optimized construction plan to construct a complete process network diagram. In this network diagram, each node represents a process, and the lines between nodes represent the dependencies between processes. In order to deal with the uncertainty that may be brought about by special processes, the PERT (Project Review and Evaluation Technique) method is used to estimate the most optimistic, most likely, and most pessimistic durations for each process, thereby obtaining the probability distribution of the construction period. Then, the critical path algorithm (CPM) can be applied to identify the critical process sequence that affects the total construction period. During the calculation process, special attention should be paid to the impact of special processes on other processes. For example, the reserved column process may affect the start time of subsequent structural construction, the reverse loading process may affect the progress of surrounding construction activities, and the water injection process of prefabricated steel blocks may affect the timing of waterproof layer construction. By running Monte Carlo simulation, the probability distribution of the project completion time is obtained, and the special process with the greatest impact on the total construction period is identified. Based on these analysis results, an adjustable schedule can be formulated, which not only includes the arrangement of conventional processes, but also takes into account the time window and resource requirements of special processes. In order to improve the flexibility of the plan, the concept of buffer time is introduced, and appropriate time buffers are set before and after special processes on the critical path to deal with possible delays or technical problems. In addition, resource balancing technology is used to ensure that the professional equipment and personnel required for special processes can be reasonably allocated. By combining this adjustable schedule with a real-time monitoring system, dynamic adjustment of the construction progress can be achieved, and changes in the actual situation on site can be responded to in a timely manner, thereby ensuring that the entire subway station construction project can be completed on time and with high quality.

[0082] Step S105: Based on the adjustable progress plan, the construction process of the subway station is monitored and analyzed in real time to obtain a real-time state parameter set, wherein the real-time state parameter set includes: surface settlement state, surrounding inclination state and soil stress state;

[0083] Specifically, various sensors and monitoring equipment are deployed at the construction site, such as high-precision levels for monitoring surface settlement, inclinometers for monitoring the inclination of surrounding buildings, and earth pressure gauges and strain gauges for monitoring soil stress state. These devices should correspond to the key nodes and processes in the adjustable schedule to ensure that accurate monitoring data can be obtained at every important stage of construction. For example, when excavating deep foundation pits, the layout of surface settlement monitoring points may be intensified, and the frequency of inclination monitoring of surrounding buildings may be increased. At the same time, a real-time data transmission and processing system is established to transmit the data collected on site to the central control room in real time, which can be achieved through wireless sensor networks or fiber-optic communication systems. In terms of data processing, signal processing technology is used to filter and denoise the raw data to eliminate the influence of environmental interference and measurement errors. Next, data fusion technology, such as Kalman filtering or Bayesian estimation, is used to comprehensively process the monitoring data from different sources to obtain a more accurate and reliable state estimation. In order to achieve real-time analysis of the construction process, an AI-based early warning system can be developed, which can predict the changing trends of surface settlement, surrounding inclination and soil stress in real time based on historical data and current monitoring data. For example, a long short-term memory network (LSTM) is used to establish a time series prediction model, or Gaussian process regression is used for spatial interpolation and prediction. These prediction results can be compared with the preset safety threshold, and once an abnormality is found, the system will automatically issue an early warning signal. In addition, the real-time monitoring data can be associated with the adjustable schedule to evaluate the impact of the current construction schedule on the surface settlement, surrounding inclination and soil stress state, and the construction plan can be dynamically adjusted according to the analysis results. For example, if the surface settlement rate in a certain area is monitored to exceed expectations, it may be necessary to adjust the excavation sequence or strengthen the support measures. In this way, not only can a real-time state parameter set containing the surface settlement state, the surrounding inclination state and the soil stress state be obtained, but also the construction process can be dynamically optimized and controlled based on these parameters, thereby ensuring the safety and efficiency of subway station construction.

[0084] Step S106: based on the real-time status parameter set, the adjustable progress plan corresponding to the subway station expansion construction process is updated to obtain an updated progress plan and control measures data, wherein the control measures data include active precipitation coupled deformation control measures and station top section dynamic loading control measures.

[0085] Specifically, the real-time state parameter set is compared with the preset safety threshold and construction target to evaluate the degree of deviation of the current construction state. For example, if the surface settlement state exceeds the warning value, or the inclination of the surrounding buildings approaches the critical value, the system will trigger the update process of the schedule. During the update process, dynamic programming algorithms or heuristic algorithms are used to re-optimize the construction sequence and resource allocation to cope with the current construction risks. At the same time, considering the trade-off between construction progress and safety, it may be necessary to extend the duration of certain key processes or add additional safety measures. For example, if the soil stress state is abnormal, it may be necessary to adjust the excavation sequence or increase the support strength, and these adjustments will be reflected in the updated schedule. When formulating the control measure data, the active precipitation coupled deformation control measures are first considered. This involves establishing a coupled system of a hydrogeological model and a formation deformation model, and predicting the impact of precipitation on formation deformation through numerical simulation. Based on the simulation results, the layout of precipitation wells and the pumping strategy can be optimized, such as adjusting the depth, spacing and pumping rate of precipitation wells to maximize the control of formation deformation. At the same time, the impact of precipitation on the surrounding environment is also considered, and a recharge system may need to be designed to maintain groundwater balance. For the dynamic loading control measures in the station top interval, it is necessary to establish a mechanical model that considers soil-structure interaction, and simulate the impact of different loading schemes on the subway station structure and surrounding strata through finite element analysis or boundary element method. Based on the analysis results, a dynamic loading scheme can be designed, such as controlling stratum deformation and structural stress by adjusting the loading sequence, loading rate and loading intensity. During the implementation process, a real-time feedback control strategy is adopted to dynamically adjust the loading parameters according to the monitoring data to ensure the safety and effectiveness of the loading process. Finally, the updated schedule needs to be integrated with the control measures data to form a comprehensive construction management plan. This plan not only includes the adjusted process arrangement and resource allocation, but also details the specific implementation steps and parameter settings of active precipitation and dynamic loading. In this way, dynamic optimization and refined control of the subway station construction process can be achieved, various risks and challenges in the construction process can be effectively dealt with, and the project can be completed safely and efficiently.

[0086] In the embodiment of the present application, by collecting and preprocessing the multi-source data related to the subway station during the bilateral expansion construction process, a standardized data set including design parameters, monitoring data and construction management data is obtained, which provides a reliable basis for subsequent analysis. These data are subjected to multi-dimensional fusion analysis to obtain target evaluation data including earthwork excavation evaluation data, support system evaluation data and deformation control evaluation data, so as to achieve comprehensive consideration of key factors. Based on these evaluation data, the preset construction plan is intelligently optimized to obtain an optimized construction plan including layered and block excavation, back pressure island strip extraction and early closure construction sub-plans, which improves construction efficiency and reduces risks. The optimization plan is analyzed by the critical path algorithm, and special processes such as reserved columns, reverse thrust loading and steel prefabricated block water injection are considered to obtain an adjustable schedule, which improves the accuracy and executability of the plan. Based on this plan, the construction process is monitored and analyzed in real time, and a real-time state parameter set including surface settlement state, surrounding inclination state and soil stress state is obtained, which provides a basis for timely discovery of problems. Finally, the progress plan was dynamically updated based on these real-time data, and a control plan was formulated, including active precipitation coupled deformation control and station top interval dynamic loading control measures, to achieve flexible adjustment of the construction plan. This method achieves comprehensive monitoring, accurate evaluation and dynamic optimization of the construction process, significantly improves the efficiency and accuracy of construction management, reduces risks, improves quality, and provides strong support for the safe and efficient construction of subway stations.

[0087] In a specific embodiment, the process of executing step S101 may specifically include the following steps:

[0088] (1) The design drawings of the subway station are digitized to obtain a three-dimensional structural model including reserved columns, prefabricated steel blocks, and internal buttress diaphragm walls. The three-dimensional structural model is then parametrically decomposed to obtain a design parameter set that reflects the structural characteristics of the subway station.

[0089] (2) Real-time monitoring of the subway station construction site is performed through a distributed sensor network to obtain the original monitoring data stream, and noise is eliminated from the original monitoring data stream to obtain a cleaned monitoring data set;

[0090] (3) Perform spatiotemporal alignment processing on the cleaned monitoring data set to obtain multidimensional monitoring data in a unified coordinate system, and digitally record the personnel, equipment, and material information at the construction site to obtain the original construction management data;

[0091] (4) Structural processing is performed on the original construction management data to obtain a construction management database containing process dependencies, and the design parameter set, multi-dimensional monitoring data, and construction management database are processed in a unified format to obtain a preliminary standardized data set;

[0092] (5) Perform data cleaning on the preliminary standardized data set to obtain an intermediate data set that removes outliers, and perform data completion and normalization on the intermediate data set to obtain a standardized data set.

[0093] Specifically, computer-aided design (CAD) software and building information modeling (BIM) technology are used to digitize the design drawings of subway stations. By converting two-dimensional drawings into three-dimensional models, key structural elements such as reserved columns, prefabricated steel blocks, and internal buttress ties can be accurately represented. On this basis, parametric design methods can be used to convert each structural element in the three-dimensional model into quantifiable parameters, such as the height and diameter of reserved columns, the size and spacing of prefabricated steel blocks, etc. These parameters together constitute a design parameter set that reflects the structural characteristics of the subway station.

[0094] In order to achieve real-time monitoring of the construction site, a distributed sensor network needs to be deployed. This network may include various types of sensors such as strain gauges, inclinometers, displacement sensors, etc., which are distributed at key locations on the construction site. These sensors transmit data in real time through wireless communication technology to form a raw monitoring data stream. However, the raw data often contains noise and needs to be denoised. Methods such as wavelet transform or Kalman filtering are used to eliminate noise and obtain a cleaned monitoring data set. For example, the process of denoising using wavelet transform can be expressed as:

[0095] ;

[0096] in, is a noisy signal, It is a real signal. is noise. Through wavelet transform and threshold processing, it can effectively separate , and get the cleaned signal.

[0097] Temporal and spatial alignment of the cleaned monitoring data is a key step to ensure data consistency. This involves coordinate system conversion and time synchronization. You can choose a unified coordinate system (such as WGS84) and use a coordinate conversion algorithm to convert all spatial data into this unified system. Time synchronization can be achieved through the Network Time Protocol (NTP). After this processing, multi-dimensional monitoring data in a unified coordinate system is obtained.

[0098] At the same time, the personnel, equipment and material information at the construction site also needs to be digitally recorded. This can be achieved by using mobile applications, RFID technology or QR code scanning. These data constitute the original construction management data. These data are structured and stored using relational databases or graph databases. In particular, the process dependency can be represented by a directed graph, thus forming a construction management database containing process dependency.

[0099] In order to integrate the design parameter set, multi-dimensional monitoring data and construction management database, the data format needs to be standardized. This involves operations such as data type conversion, unit unification, and timestamp formatting. A common data exchange format, such as JSON or XML, can be selected to represent all data, thereby obtaining a preliminary standardized data set.

[0100] Finally, data cleaning of the initially standardized dataset is an important step to ensure data quality. This includes identifying and handling outliers, using statistical methods such as the Z-score or interquartile range (IQR) method to detect outliers. For example, the formula for detecting outliers using the Z-score method is:

[0101] ;

[0102] in, is the observed value, is the average value, is the standard deviation. Usually, when , the observation can be considered an outlier.

[0103] For missing data, interpolation or machine learning methods are used to complete the data. Common methods include linear interpolation, polynomial interpolation, or similarity-based collaborative filtering methods. Finally, in order to make data of different dimensions comparable, normalization is required. Common normalization methods include minimum-maximum normalization or Z-score normalization.

[0104] In a specific embodiment, the process of executing step S102 may specifically include the following steps:

[0105] (1) Extract the features of the design parameters in the standardized data set to obtain the structural feature vector of the subway station, and perform spatiotemporal correlation analysis on the monitoring data to obtain a multidimensional monitoring data correlation matrix;

[0106] (2) The construction management data is decomposed into process steps to obtain the sequence of key nodes in the construction process. Based on the structural characteristic vector of the subway station, the earthwork excavation scheme is parametrically modeled to obtain the earthwork excavation numerical model.

[0107] (3) Simulate the stress field of the earthwork excavation numerical model to obtain earthwork excavation evaluation data, and dynamically evaluate the stress state of the support system based on the multi-dimensional monitoring data correlation matrix to obtain support system evaluation data;

[0108] (4) Combined with the sequence of key nodes in the construction process, the deformation monitoring data is trend analyzed to obtain a deformation development prediction model, and the deformation development prediction model is classified into risk levels to obtain deformation control assessment data;

[0109] (5) Multi-criteria decision analysis is performed on the earthwork excavation assessment data, support system assessment data, and deformation control assessment data to obtain comprehensive target evaluation indicators. Based on the comprehensive target evaluation indicators, a multi-target evaluation system for subway station construction progress, safety, and quality is constructed to obtain target evaluation data.

[0110] Specifically, when extracting features from the design parameters in the standardized data set, dimensionality reduction techniques such as principal component analysis (PCA) or autoencoders are used to obtain a low-dimensional vector representation that reflects the key structural features of the subway station. This structural feature vector may contain key information such as the depth of the station hall, the type of enclosure structure, and geological conditions. At the same time, when analyzing the spatiotemporal correlation of the monitoring data, time series analysis methods such as autocorrelation function (ACF) and cross-correlation function (CCF) are used to reveal the spatiotemporal correlation between different monitoring points. These analysis results can be organized into a multidimensional monitoring data correlation matrix, in which the matrix elements represent the correlation coefficients between different monitoring points.

[0111] Decomposing the construction management data into process steps is a key step in building a construction process model. Using the work breakdown structure (WBS) method in project management, the entire construction process is decomposed into a series of specific work packages, and the key node sequence of the construction process is determined by the critical path method (CPM) or the program evaluation and review technique (PERT). This sequence provides a time frame for subsequent progress analysis and risk assessment. Based on the previously obtained subway station structural characteristic vector, the earthwork excavation plan can be parametrically modeled. Here, numerical simulation methods such as the finite element method (FEM) or discrete element method (DEM) are used to construct a numerical model that reflects the mechanical behavior of the soil. The input parameters of the model include soil layer parameters, excavation depth, support structure parameters, etc., which can be extracted from the structural characteristic vector.

[0112] With the numerical model of earth excavation, the stress field simulation can be performed. This usually involves solving a complex set of partial differential equations. For example, in the framework of elastic mechanics, the governing equation of the stress field can be expressed as:

[0113] ;

[0114] in, is the stress tensor, is the volume force, is the density, is the displacement vector, Denotes the divergence operator. By solving this set of equations, the stress distribution of the soil during excavation is obtained, thereby evaluating the safety and stability of the excavation.

[0115] At the same time, based on the multi-dimensional monitoring data association matrix, the stress state of the support system can be dynamically evaluated. This can be achieved by building a structural mechanics model and using the real-time monitoring data as input. For example, for a complex support system, its dynamic response can be described by the following state space equation:

[0116] ;

[0117] ;

[0118] in, is the system state vector (including displacement, velocity, etc.), is the input vector (such as external load), is the output vector (such as strain, stress, etc.), A, B, C are system matrices, and They are process noise and measurement noise. Through methods such as Kalman filtering, the system state can be estimated in real time to evaluate the stress condition of the support system.

[0119] Combined with the sequence of key nodes in the construction process, the deformation monitoring data can be trend analyzed. Time series analysis methods, such as ARIMA model or long short-term memory network (LSTM) and other deep learning methods, are used to build a deformation development prediction model. Based on the prediction results, multiple risk thresholds can be set to divide the deformation development process into different risk levels, thereby obtaining deformation control assessment data.

[0120] Finally, in order to comprehensively consider the evaluation results of the three aspects of earthwork excavation, support system and deformation control, a multi-criteria decision analysis is required. This can be achieved through the analytic hierarchy process (AHP) or the fuzzy comprehensive evaluation method. Through expert scoring and weight calculation, a comprehensive target evaluation index is obtained. Based on this index, a multi-objective evaluation system for subway station construction progress-safety-quality can be constructed. This evaluation system can be expressed as a multi-objective optimization problem:

[0121] ;

[0122] in, , , , represent the progress, safety and quality objective functions respectively, g(x) and h(x) is a constraint, Xis the feasible domain of decision variables. By solving this multi-objective optimization problem, a set of optimal solutions that balance progress, safety, and quality are obtained, which constitute the final target evaluation data.

[0123] Through this series of complex analysis and modeling processes, the transformation from raw data to comprehensive evaluation results is achieved, providing a scientific basis for intelligent management and decision-making of subway station construction. This method not only takes into account multiple key aspects of the construction process, but also realizes quantitative evaluation and prediction of complex systems through mathematical models and data analysis technology, thereby greatly improving the accuracy and reliability of construction management.

[0124] In a specific embodiment, the process of executing step S103 may specifically include the following steps:

[0125] (1) Decompose the target evaluation data into multiple targets to obtain the weight coefficients of earthwork excavation target, support system target, and deformation control target. Based on the weight coefficients, make a preliminary score for the preset construction plan to obtain the initial score matrix.

[0126] (2) The initial scoring matrix is ​​optimized and iterated through a genetic algorithm to obtain a set of candidate solutions that meet the multi-objective requirements. The layered and block excavation schemes in the candidate solution set are parameterized and modeled to obtain a layered and block model that reflects the excavation depth, block size, and excavation sequence.

[0127] (3) Based on the layered and block model, the geological stress field is simulated to obtain the stress distribution data during the layered and block excavation process. The layered and block excavation scheme is optimized based on the stress distribution data to obtain the layered and block excavation construction sub-scheme;

[0128] (4) Geometric optimization is performed on the anti-pressure island stripping schemes in the candidate scheme set to obtain the optimal anti-pressure island shape and stripping width parameters. Based on the optimal anti-pressure island shape and stripping width parameters, stability analysis is performed to obtain the safety factor curve of the anti-pressure island stripping construction;

[0129] (5) The anti-pressure island stripping scheme is optimized through the safety factor curve to obtain the anti-pressure island stripping construction sub-scheme;

[0130] (6) Optimize the process of the early closure scheme in the candidate set to obtain the optimal closure time and closure sequence, and perform deformation prediction analysis based on the optimal closure time and closure sequence to obtain deformation control effect evaluation data of early closure construction;

[0131] (7) Optimize the early closure scheme based on the deformation control effect evaluation data to obtain the early closure construction sub-scheme;

[0132] (8) The data of the layered and block excavation construction sub-scheme, the counter-pressure island strip extraction construction sub-scheme, and the early closure construction sub-scheme were integrated to obtain an optimized construction plan.

[0133] Specifically, when multi-objective decomposition of target evaluation data is performed, the analytic hierarchy process (AHP) or fuzzy comprehensive evaluation method is used to determine the weight coefficients of earthwork excavation targets, support system targets, and deformation control targets. This process involves constructing a judgment matrix, calculating eigenvectors, and performing consistency checks. For example, when using the AHP method, a 3x3 judgment matrix A can be constructed as follows:

[0134]

[0135] in Indicates the target Relative to target The importance of then solving the characteristic equation ,in, is the identity matrix, is the maximum eigenvalue, is the corresponding eigenvector, which is the required weight coefficient.

[0136] With the weight coefficients, the preset construction plans can be initially scored. Assuming there are m preset plans and n evaluation indicators, an m×n initial scoring matrix R is obtained. In order to optimize this scoring matrix, a genetic algorithm is used. The key to the genetic algorithm is to design a suitable chromosome encoding method, fitness function and genetic operator. Here, each chromosome can represent a construction plan, and the fitness function can be defined as a weighted objective function:

[0137] ;

[0138] in , , is the weight coefficient obtained before, , , They are the objective functions of earthwork excavation, support system and deformation control. Through multiple generations of iteration, a set of candidate solutions that meet the multi-objective requirements is obtained.

[0139] The layered and block excavation schemes in the candidate set are parametrically modeled. This can be achieved by building a three-dimensional geometric model, which contains parameters such as excavation depth, block size and excavation sequence. Based on this layered and block model, the geological stress field simulation can be carried out. This usually involves solving a complex set of elastic-plastic mechanical equations. For example, under the small deformation assumption, the stress-strain relationship can be expressed as:

[0140] ;

[0141] in is the stress tensor, The elastic constant tensor, is the total strain tensor, is the plastic strain tensor. By solving this set of equations using the finite element method, we can obtain the stress distribution data during the layered and block excavation process. Based on these data, the layered and block excavation scheme can be further optimized, such as adjusting the excavation sequence or support measures to minimize stress concentration or the expansion of the plastic zone.

[0142] For the anti-pressure island stripping scheme in the candidate solution set, geometric optimization is required to determine the optimal anti-pressure island shape and stripping width parameters. This can be achieved through parametric design and topology optimization methods. For example, the level set method is used to describe the shape of the anti-pressure island. After the optimal shape is obtained, stability analysis can be performed to obtain the safety factor curve of the anti-pressure island stripping construction, and the construction plan can be further optimized based on this.

[0143] For the early closure solution, process optimization is required to determine the optimal closure timing and order. This can be achieved by establishing a discrete event simulation model, which takes into account the dependencies between the processes and resource constraints. Through Monte Carlo simulation or heuristic algorithms, an optimal solution that balances construction period and resource utilization can be found. Based on this optimal solution, deformation prediction analysis can be performed, such as using the finite difference method to solve the governing equations of foundation deformation:

[0144] ;

[0145] in is the displacement, is the wave speed, is the external force term, represents the spatial coordinate variable (indicating the position), Represents a time variable.

[0146] By solving this equation, the deformation control effect evaluation data of early closure construction is obtained, and the early closure scheme is further optimized based on this.

[0147] Finally, data fusion is performed on the layered and block excavation construction sub-scheme, the counter-pressure island stripping construction sub-scheme, and the early closure construction sub-scheme to obtain the final optimized construction plan. This can be achieved through multi-criteria decision-making methods, such as using fuzzy comprehensive evaluation method or grey correlation analysis method. By constructing an evaluation index system, determining the index weights, and calculating the comprehensive evaluation value, an optimized construction plan that fully considers all factors is obtained. Taking into account multiple key aspects of the construction process, the precise solution and optimization of complex engineering problems are achieved, thereby greatly improving the scientificity and feasibility of the construction plan. This method based on the combination of data-driven and model-driven provides strong support for intelligent decision-making in complex construction environments of subway stations, helps to improve construction efficiency, ensure construction safety, and ultimately achieve an overall improvement in project quality.

[0148] In a specific embodiment, the process of executing step S104 may specifically include the following steps:

[0149] (1) The reserved column process in the special process database is parameterized to obtain the key parameter set of the reserved column construction. Based on the key parameter set, the reserved column construction process is decomposed to obtain the detailed process network of the reserved column construction;

[0150] (2) Perform a time series analysis on the reverse loading process in the special process database to obtain the optimal implementation time of reverse loading. Based on the optimal implementation time, optimize the resource allocation of the reverse loading process to obtain the resource demand curve of reverse loading.

[0151] (3) Based on the resource demand curve of the reverse loading, a process analysis is performed on the water injection process of the prefabricated steel blocks in the reverse loading process to obtain the quality control points of the water injection of the prefabricated steel blocks, and based on the quality control points, a time estimation is performed on the water injection process of the prefabricated steel blocks to obtain the duration distribution of the water injection process;

[0152] (4) Integrate the process steps of the layered and block excavation construction sub-scheme, the counter-pressure island strip extraction construction sub-scheme, and the early closure construction sub-scheme in the optimized construction scheme to obtain a comprehensive construction network diagram;

[0153] (5) The comprehensive construction network diagram is calculated using the critical path algorithm to obtain an adjustable schedule.

[0154] Specifically, the realization of this complex special process optimization and schedule planning process requires the comprehensive use of a variety of advanced data processing, optimization algorithms and project management technologies. First, when parameterizing the reserved column process in the special process database, data mining and feature extraction techniques are used to extract key parameters from historical data, such as the height, diameter, reinforcement configuration, and concrete strength of the reserved column. These parameters can be reduced in dimension by methods such as principal component analysis (PCA) or autoencoders to obtain a low-dimensional set of key parameters. Based on this set of key parameters, the work breakdown structure (WBS) method is used to decompose the reserved column construction process to obtain a refined process network. This network can be represented by a directed acyclic graph (DAG), in which nodes represent processes and edges represent dependencies between processes. When performing time series analysis on the reverse loading process, time series analysis methods are used, such as deep learning methods such as the autoregressive integrated moving average (ARIMA) model or the long short-term memory network (LSTM). Assuming the ARIMA (p, d, q) model is used, where p is the number of autoregressive terms, d is the number of differences, and q is the number of moving average terms, the model can be expressed as:

[0155] ;

[0156] in is the lag operator, is an autoregressive polynomial, is the moving average polynomial, is a time series, is white noise. By fitting this model, the optimal implementation time of reverse loading can be predicted. Based on this optimal time, a resource balancing algorithm, such as the heuristic algorithm of the resource-constrained project scheduling problem (RCPSP), is used to optimize the resource allocation of the reverse loading process. This will produce a resource demand curve that reflects the demand for manpower, equipment, and materials.

[0157] For the water injection process of steel precast blocks, a detailed process analysis is required. This can be done by identifying key quality control points through methods such as failure mode and effect analysis (FMEA) or quality function deployment (QFD). Each quality control point can be represented by a triple (S, O, D), where S is the severity, O is the frequency of occurrence, and D is the difficulty of detection. The risk priority number (RPN) can be calculated by the following formula:

[0158] ;

[0159] The most critical quality control points can be determined based on the RPN value. Based on these control points, the PERT technique is used to estimate the duration of the water injection process. PERT assumes that the duration of each activity follows a Beta distribution, and its expected value E and variance V can be calculated by the following formula:

[0160] ;

[0161] ;

[0162] Among them, a is the most optimistic time estimate, m is the most likely time estimate, and b is the most pessimistic time estimate. Through MonteCarlo simulation, the duration distribution of the water injection process is obtained. After obtaining the optimization results of these special processes, they need to be integrated with the previously optimized layered and block excavation construction sub-scheme, the back pressure island strip drawing construction sub-scheme, and the early closure construction sub-scheme. This can be achieved by constructing a comprehensive project network diagram. Use graph merging algorithms in graph theory, such as the Union Graph method, to merge the network graphs of each sub-scheme into a large comprehensive construction network graph. In this process, special attention should be paid to handling the interfaces and dependencies between different sub-schemes. Perform a critical path analysis on this comprehensive construction network diagram. The critical path method (CPM) is a commonly used project scheduling method that can help identify a series of activities that have the greatest impact on the total project duration. Assuming that the network diagram has n activities, each activity i has a duration di, the following variables can be defined:

[0163] ES(i): the earliest start time of activity i

[0164] EF(i): earliest completion time of activity i

[0165] LS(i): the latest start time of activity i

[0166] LF(i): latest completion time of activity i

[0167] The core of the critical path algorithm is to solve the following set of equations:

[0168] EF(i) = ES(i) + di

[0169] ES(j) = max{EF(i) | i is the direct predecessor activity of j}

[0170] LF(i) = min{LS(j) | j is the direct successor activity of i}

[0171] LS(i) = LF(i) - di.

[0172] By solving this set of equations, the earliest and latest start / finish time of each activity is obtained, thereby identifying the critical path and key activities. This result constitutes an adjustable schedule. It realizes the optimization of details of special processes and the formulation of the overall schedule.

[0173] In a specific embodiment, the process of executing step S105 may specifically include the following steps:

[0174] (1) The adjustable schedule is divided into time windows to obtain the key monitoring time node sequence of the construction process. Based on the key monitoring time node sequence, the data collection frequency of the sensor network at the subway station construction site is optimized to obtain a dynamic sampling scheme.

[0175] (2) Perform wavelet transform noise reduction on the collected original monitoring data to obtain cleaned surface settlement data, cleaned surrounding inclination data, and cleaned soil stress data;

[0176] (3) Perform spatiotemporal interpolation on the cleaned surface subsidence data to obtain a continuous surface subsidence field;

[0177] (4) Based on the surface subsidence field, the curvature analysis algorithm is used to identify abnormal subsidence areas and obtain the surface subsidence status assessment results;

[0178] (5) Perform trend analysis on the surrounding inclination data after cleaning to obtain the building inclination rate curve. Based on the building inclination rate curve, the stability level of the surrounding buildings is determined by the threshold discrimination method to obtain the surrounding inclination status assessment result;

[0179] (6) Perform stress path analysis on the soil stress data after cleaning to obtain the stress evolution process of key parts. Based on the stress evolution process, calculate the soil stability index through the plastic zone expansion prediction model to obtain the soil stress state assessment result;

[0180] (7) The surface settlement state assessment results, the surrounding inclination state assessment results, and the soil stress state assessment results are fused to obtain a real-time state parameter set.

[0181] Specifically, when the adjustable schedule is divided into time windows, the dynamic time warping (DTW) algorithm is used to identify key construction stages and turning points, thereby obtaining a sequence of key monitoring time nodes for the construction process. This sequence not only reflects the important milestones of the construction progress, but also takes into account the different monitoring requirements of different construction stages. Based on this sequence of key monitoring time nodes, the data collection frequency of the sensor network at the subway station construction site can be optimized. This can be achieved through an adaptive sampling algorithm, such as using an extended form of Shannon's sampling theorem:

[0182] ;

[0183] Where fs is the sampling frequency, B is the highest frequency component of the signal, and α is a safety factor used to deal with non-ideal low-pass filter characteristics. By dynamically adjusting the α value, the sampling frequency can be flexibly changed at different construction stages, thereby obtaining a dynamic sampling scheme that balances data accuracy and transmission efficiency. Performing wavelet transform denoising on the collected raw monitoring data is a key step in improving data quality. Discrete wavelet transform (DWT) is used to achieve this goal. Assuming the original signal is f(t), its wavelet transform can be expressed as:

[0184] ;

[0185] in is the scale parameter, is the translation parameter, is the wavelet mother function, Represents complex conjugate. The wavelet coefficients are processed by selecting appropriate thresholds and then inversely transformed to obtain the cleaned surface settlement data, surrounding inclination data and soil stress data.

[0186] For the cleaned surface settlement data, spatiotemporal interpolation is required to obtain a continuous surface settlement field. This can be achieved through the Kriging interpolation method, which takes into account spatial autocorrelation and can provide the best linear unbiased estimate (BLUE). The basic formula of Kriging interpolation is:

[0187] ;

[0188] in is the predicted value of the point to be estimated, is the observed value of the known point, is the weight coefficient, which needs to be determined by analyzing the variogram and solving the Kriging equations. In this way, a high-precision continuous surface settlement field is obtained. Based on this surface settlement field, the curvature analysis algorithm is used to identify the settlement anomaly area. Specifically, the Laplace operator of the settlement field can be calculated:

[0189] ;

[0190] in is the sedimentation field function, is the field function to be analyzed (here it represents the surface deformation or settlement field); is the horizontal coordinate in the plane rectangular coordinate system (indicating the position in the east-west direction); It is the ordinate in the plane rectangular coordinate system (indicating the position in the north-south direction). By setting an appropriate threshold, the area with abnormal curvature can be identified, thereby obtaining the surface settlement status assessment result.

[0191] For the cleaned perimeter inclination data, trend analysis is required to obtain the building inclination rate curve. This can be achieved through time series analysis methods, such as using an exponential smoothing model:

[0192] ;

[0193] in yes The smoothed value at time, is the actual observed value, is the smoothing coefficient (0< <1). In this way, a smooth curve reflecting the tilt trend is obtained. Based on this tilt rate curve, the threshold discrimination method is used to determine the stability level of the surrounding buildings, thereby obtaining the surrounding tilt state assessment result. For the soil stress data after cleaning, stress path analysis is required to obtain the stress evolution process of key parts. This can be achieved through stress invariant analysis, such as calculating the deviatoric stress q and the average principal stress p:

[0194] ;

[0195] ;

[0196] in , , is the principal stress. qp The stress state of the soil can be evaluated by using the path. Based on this stress evolution process, the soil stability index is calculated using the plastic zone expansion prediction model. For example, using the Mohr-Coulomb yield criterion:

[0197] ;

[0198] in φ is the internal friction angle, c is the cohesion. By calculating the expansion speed and range of the plastic zone, the soil stress state assessment results are obtained. Finally, the surface settlement state assessment results, the surrounding inclination state assessment results and the soil stress state assessment results are fused to obtain a comprehensive real-time state parameter set. This can be achieved through a multi-sensor data fusion algorithm, such as using a Kalman filter or a Bayesian network. Assuming the use of a Kalman filter, its basic equation is:

[0199] ;

[0200] ;

[0201] in yes The state estimate at time is the state transition matrix, is the control input matrix, is the control vector, is the process noise, is the observation vector, is the observation matrix, is the observation noise, yes The actual state vector at the moment. By recursively applying the prediction and update steps, an optimal state estimate that comprehensively considers various monitoring data is obtained, that is, a set of real-time state parameters.

[0202] In a specific embodiment, the process of executing step S106 may specifically include the following steps:

[0203] (1) Evaluate the risk level of the real-time status parameter set to obtain the current risk index of the subway station construction. Based on the current risk index, perform deviation analysis on the adjustable schedule to obtain the schedule deviation matrix.

[0204] (2) Perform decision analysis on the schedule deviation matrix to obtain a schedule adjustment strategy set. Based on the schedule adjustment strategy set, optimize the time parameters of the adjustable schedule to obtain a preliminary updated schedule.

[0205] (3) Re-allocate resources for the preliminary update schedule to obtain an updated schedule after resource balance. Based on the updated schedule, optimize the parameters of the active precipitation system to obtain the precipitation well layout plan and pumping rate curve.

[0206] (4) The dewatering well layout scheme and the pumping rate curve were coupled and analyzed to obtain the active dewatering coupled deformation control measures. Based on the updated schedule, the station top section was loaded with simulation to obtain the dynamic loading force distribution diagram.

[0207] (5) The dynamic loading force distribution diagram is optimized in time series to obtain the dynamic loading control measures for the station top interval. The active precipitation coupled deformation control measures and the dynamic loading control measures for the station top interval are integrated into a control measure data package.

[0208] Specifically, realizing this complex risk assessment and control measures optimization process requires the comprehensive use of a variety of advanced data analysis, decision optimization and numerical simulation technologies. First, when evaluating the risk level of the real-time state parameter set, the fuzzy comprehensive evaluation method is used. Assuming there are n evaluation indicators and m risk levels, the following fuzzy comprehensive evaluation model can be established:

[0209] ;

[0210] in is the final risk assessment result vector, is the weight vector, It is The membership vector of each indicator. Through this model, a comprehensive risk index is obtained to reflect the current risk status of the subway station construction. Based on this risk index, the deviation analysis of the adjustable schedule can be performed. This can be achieved by calculating the difference between the actual schedule and the planned schedule, and obtaining a schedule deviation matrix D.

[0211] When performing decision analysis on the schedule deviation matrix, a multi-criteria decision method is used, such as the analytic hierarchy process (AHP) or the TOPSIS method. Assuming that the TOPSIS method is used, its core idea is to select the solution that is closest to the ideal solution and farthest from the negative ideal solution. A weighted normalized decision matrix V can be defined, whose elements are The calculation is as follows:

[0212] ;

[0213] in, It is The weight of the criteria, It is The scheme is in By calculating the distance between each solution and the ideal solution and the negative ideal solution, a schedule adjustment strategy set is obtained. Based on this strategy set, the time parameters of the adjustable schedule can be optimized, which can be achieved through the critical path method (CPM) or project evaluation and review technique (PERT) to obtain a preliminary updated schedule.

[0214] For this initial updated schedule, resource reallocation is required to ensure efficient resource utilization. This can be achieved through resource balancing algorithms, such as heuristic algorithms for the Resource Constrained Project Scheduling Problem (RCPSP). Assume that resources, each resource In time The available amount is ,Activity The duration is , the resource demand is , then the resource constraint can be expressed as:

[0215] ;

[0216] in Is a 0-1 variable indicating activity Is it in time Start. By solving this optimization problem, we get an updated schedule after resource balancing.

[0217] Based on this updated schedule, the parameters of the active dewatering system are optimized. This can be achieved by building a groundwater flow model, such as using MODFLOW software for numerical simulation, assuming a three-dimensional groundwater flow equation under steady flow conditions, and solving this equation to obtain the optimal dewatering well layout and pumping rate curve.

[0218] When coupling the dewatering well arrangement and the pumping rate curve, the effect of dewatering on formation deformation is considered. This can be achieved by establishing a fluid-solid coupling model, for example using Biot consolidation theory:

[0219] ;

[0220] ;

[0221] in is the total stress tensor, is the density, is the acceleration due to gravity, is the porosity, is the seepage velocity, is the source-sink term. By solving this coupled set of equations, the active precipitation coupled deformation control measures are obtained. At the same time, based on the updated schedule, the loading simulation of the station top interval is carried out. This can be achieved through finite element analysis, such as dynamic analysis using ABAQUS software. A model considering soil-structure interaction can be established, and its control equation can be expressed as:

[0222] ;

[0223] Where M, C, and K are the mass matrix, damping matrix, and stiffness matrix, respectively. is the displacement vector, is the time-varying load vector, is the velocity vector, is the first-order derivative of displacement with respect to time, is the acceleration vector, is the second-order derivative of displacement with respect to time, and by solving this equation, the dynamic loading force distribution diagram is obtained.

[0224] The dynamic loading force distribution diagram is optimized in time sequence using a dynamic programming algorithm. Assuming there are n time steps and m possible loading schemes for each time step, the optimal loading sequence can be solved by the following recursive formula:

[0225] ;

[0226] in Before time steps, the last step is to choose a solution The minimum cost when Indicates that from the plan Switch to Scheme The cost of the optimization problem is . By solving this optimization problem, the dynamic loading control measures for the station top interval are obtained. The active precipitation coupled deformation control measures and the station top interval dynamic loading control measures are integrated into a unified control measure data package. This is achieved by establishing a comprehensive control model that takes into account the mutual influence between precipitation, deformation and loading. This data package will provide comprehensive guidance for real-time control of the construction site, help ensure construction safety and improve construction efficiency.

[0227] The above describes the design and construction scheme optimization method for subway double-sided extension and heave control in the embodiment of the present application. The following describes the design and construction scheme optimization system for subway double-sided extension and heave control in the embodiment of the present application. Please refer to Figure 2 , an embodiment of the design and construction scheme optimization system for subway double-sided expansion and construction heave and sink control in the embodiment of the present application includes:

[0228] The processing module 201 is used to collect and preprocess multi-source data related to the subway station during the bilateral extension construction process to obtain a standardized data set, wherein the standardized data set includes design parameters, monitoring data, and construction management data;

[0229] A fusion module 202 is used to perform multi-dimensional fusion analysis on the standardized data set to obtain target evaluation data, wherein the target evaluation data includes: earth excavation evaluation data, support system evaluation data and deformation control evaluation data;

[0230] The optimization module 203 is used to perform intelligent optimization processing on the preset construction plan based on the target evaluation data to obtain an optimized construction plan, wherein the optimized construction plan includes a layered and block excavation construction sub-plan, a counter-pressure island strip drawing construction sub-plan, and an early closure construction sub-plan;

[0231] The analysis module 204 is used to analyze the construction progress of the optimized construction plan through a critical path algorithm based on a preset special process database to obtain an adjustable progress plan, wherein the special process database includes: a reserved column process and a reverse loading process;

[0232] The monitoring module 205 is used to monitor and analyze the construction process of the subway station in real time based on the adjustable progress plan to obtain a real-time state parameter set, wherein the real-time state parameter set includes: surface settlement state, surrounding inclination state and soil stress state;

[0233] The updating module 206 is used to update the adjustable progress plan corresponding to the subway station expansion construction process based on the real-time status parameter set, and obtain an updated progress plan and control measures data, wherein the control measures data include active precipitation coupled deformation control measures and station top section dynamic loading control measures.

[0234] Through the collaboration of the above components, the multi-source data related to the subway station during the bilateral expansion construction process were collected and preprocessed to obtain a standardized data set including design parameters, monitoring data and construction management data, which provided a reliable basis for subsequent analysis. The data were analyzed in a multi-dimensional fusion manner to obtain target evaluation data including earthwork excavation evaluation data, support system evaluation data and deformation control evaluation data, which achieved comprehensive consideration of key factors. Based on these evaluation data, the preset construction plan was intelligently optimized to obtain an optimized construction plan including layered and block excavation, back pressure island stripping and early closure construction sub-plans, which improved construction efficiency and reduced risks. The progress of the optimized plan was analyzed through the critical path algorithm, and special processes such as reserved columns, reverse loading and water injection of steel prefabricated blocks were considered to obtain an adjustable progress plan, which improved the accuracy and executability of the plan. Based on this plan, the construction process was monitored and analyzed in real time to obtain a real-time state parameter set including surface settlement state, surrounding inclination state and soil stress state, which provided a basis for timely discovery of problems. Finally, the progress plan was dynamically updated based on these real-time data, and a control plan was formulated, including active precipitation coupled deformation control and station top interval dynamic loading control measures, to achieve flexible adjustment of the construction plan. This method achieves comprehensive monitoring, accurate evaluation and dynamic optimization of the construction process, significantly improves the efficiency and accuracy of construction management, reduces risks, improves quality, and provides strong support for the safe and efficient construction of subway stations.

[0235] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described systems, systems and units can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0236] 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 aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A design and construction scheme optimization method for subway double-sided expansion and subsidence control, characterized in that: The design and construction scheme optimization method for subway double-sided extension and construction heave and sink control includes: Collect and preprocess multi-source data related to the subway station during the bilateral expansion construction process to obtain a standardized data set, wherein the standardized data set includes design parameters, monitoring data, and construction management data; Performing multi-dimensional fusion analysis on the standardized data set to obtain target evaluation data, wherein the target evaluation data includes: earth excavation evaluation data, support system evaluation data and deformation control evaluation data; Based on the target evaluation data, the preset construction plan is intelligently optimized to obtain an optimized construction plan, wherein the optimized construction plan includes a layered and block excavation construction sub-plan, a counter-pressure island strip drawing construction sub-plan, and an early closure construction sub-plan; Based on the target evaluation data, the preset construction plan is intelligently optimized to obtain an optimized construction plan, wherein the optimized construction plan includes a layered and block excavation construction sub-plan, a counter-pressure island strip drawing construction sub-plan, and an early closure construction sub-plan, including: Decomposing the target evaluation data into multiple targets to obtain weight coefficients of earth excavation target, support system target and deformation control target, and based on the weight coefficients, preliminarily scoring the preset construction scheme to obtain an initial scoring matrix; The initial scoring matrix is ​​optimized and iterated by a genetic algorithm to obtain a set of candidate solutions that meet multi-objective requirements, and the layered and block excavation solutions in the candidate solution set are parameterized to obtain a layered and block model that reflects the excavation depth, block size and excavation sequence; Based on the layered and block model, a geological stress field simulation is performed to obtain stress distribution data during the layered and block excavation process, and the layered and block excavation scheme is optimized by the stress distribution data to obtain the layered and block excavation construction sub-scheme; Performing geometric optimization on the back-pressure island stripping schemes in the candidate scheme set to obtain the optimal back-pressure island shape and stripping width parameters, and performing stability analysis based on the optimal back-pressure island shape and stripping width parameters to obtain a safety factor curve for the back-pressure island stripping construction; The anti-pressure island stripping scheme is optimized by using the safety factor curve to obtain the anti-pressure island stripping construction sub-scheme; Optimizing the process of the early closure scheme in the candidate set to obtain the optimal closure time and closure sequence, and performing deformation prediction analysis based on the optimal closure time and closure sequence to obtain deformation control effect evaluation data of the early closure construction; Optimizing the early closure scheme through the deformation control effect evaluation data to obtain the early closure construction sub-scheme; Data fusion is performed on the layered and block excavation construction sub-scheme, the counter-pressure island strip drawing construction sub-scheme and the early closure construction sub-scheme to obtain the optimized construction scheme; Based on a preset special process database, the construction progress of the optimized construction plan is analyzed by a critical path algorithm to obtain an adjustable progress plan, wherein the special process database includes: a reserved column process and a reverse loading process; Based on the preset special process database, the construction progress analysis of the optimized construction plan is performed through the critical path algorithm to obtain an adjustable progress plan, wherein the special process database includes: a reserved column process and a reverse loading process, including: Parameterizing the reserved column process in the special process database to obtain a key parameter set for reserved column construction, and based on the key parameter set, performing process decomposition on the reserved column construction to obtain a detailed process network for reserved column construction; Performing a time sequence analysis on the reverse loading process in the special process database to obtain the optimal implementation time of the reverse loading, and optimizing the resource allocation of the reverse loading process based on the optimal implementation time to obtain a resource demand curve of the reverse loading; Based on the resource demand curve of the reverse loading, a process analysis is performed on the water injection process of the prefabricated steel blocks in the reverse loading process to obtain the quality control points of the water injection of the prefabricated steel blocks, and based on the quality control points, a time estimation is performed on the water injection process of the prefabricated steel blocks to obtain the duration distribution of the water injection process; The process of the layered and block excavation construction sub-scheme, the counter-pressure island strip drawing construction sub-scheme and the early closure construction sub-scheme in the optimized construction scheme are integrated to obtain a comprehensive construction network diagram; Calculating the comprehensive construction network diagram by a critical path algorithm to obtain the adjustable schedule; Based on the adjustable progress plan, the construction process of the subway station is monitored and analyzed in real time to obtain a real-time state parameter set, wherein the real-time state parameter set includes: surface settlement state, surrounding inclination state and soil stress state; Based on the real-time status parameter set, the adjustable progress plan corresponding to the subway station expansion construction process is updated to obtain an updated progress plan and control measures data, wherein the control measures data include active precipitation coupled deformation control measures and station top section dynamic loading control measures.

2. The design and construction scheme optimization method for subway double-sided extension and construction uplift and sinking control according to claim 1 is characterized in that: The multi-source data related to the subway station during the bilateral extension construction process is collected and preprocessed to obtain a standardized data set, wherein the standardized data set includes design parameters, monitoring data and construction management data, including: Digitally process the design drawings of the subway station to obtain a three-dimensional structural model including reserved columns, prefabricated steel blocks and internal buttress ground-connected walls, and parametrically decompose the three-dimensional structural model to obtain a design parameter set reflecting the structural characteristics of the subway station; Real-time monitoring of the subway station construction site is performed through a distributed sensor network to obtain an original monitoring data stream, and noise is eliminated from the original monitoring data stream to obtain a cleaned monitoring data set; Performing spatiotemporal alignment processing on the cleaned monitoring data set to obtain multi-dimensional monitoring data in a unified coordinate system, and digitally recording the personnel, equipment and material information at the construction site to obtain original construction management data; Structuring the original construction management data to obtain a construction management database containing process dependencies, and unifying the data formats of the design parameter set, multi-dimensional monitoring data and construction management database to obtain a preliminary standardized data set; The preliminary standardized data set is cleaned to obtain an intermediate data set from which outliers are removed, and the intermediate data set is completed and normalized to obtain the standardized data set.

3. The design and construction scheme optimization method for subway double-sided extension and heave control according to claim 1 is characterized in that: The standardized data set is subjected to multi-dimensional fusion analysis to obtain target evaluation data, wherein the target evaluation data includes: earthwork excavation evaluation data, support system evaluation data and deformation control evaluation data, including: Performing feature extraction on the design parameters in the standardized data set to obtain a subway station structural feature vector, and performing spatiotemporal correlation analysis on the monitoring data to obtain a multi-dimensional monitoring data association matrix; Decomposing the construction management data into process steps to obtain a sequence of key nodes in the construction process, and parametrically modeling the earth excavation scheme based on the structural characteristic vector of the subway station to obtain an earth excavation numerical model; Performing stress field simulation on the earth excavation numerical model to obtain earth excavation evaluation data, and dynamically evaluating the stress state of the support system based on the multi-dimensional monitoring data association matrix to obtain support system evaluation data; Combined with the sequence of key nodes in the construction process, trend analysis is performed on the deformation monitoring data to obtain a deformation development prediction model, and risk level classification is performed on the deformation development prediction model to obtain deformation control assessment data; A multi-criteria decision analysis is performed on the earth excavation assessment data, the support system assessment data and the deformation control assessment data to obtain comprehensive target assessment indicators, and based on the comprehensive target assessment indicators, a subway station construction progress-safety-quality multi-target evaluation system is constructed to obtain the target assessment data.

4. The design and construction scheme optimization method for subway double-sided extension and heave control according to claim 1 is characterized in that: Based on the adjustable progress plan, the construction process of the subway station is monitored and analyzed in real time to obtain a real-time state parameter set, wherein the real-time state parameter set includes: surface settlement state, surrounding inclination state and soil stress state, including: The adjustable schedule is divided into time windows to obtain a key monitoring time node sequence of the construction process, and based on the key monitoring time node sequence, the data collection frequency of the sensor network at the subway station construction site is optimized to obtain a dynamic sampling scheme; The collected original monitoring data is subjected to wavelet transform noise reduction processing to obtain cleaned surface settlement data, cleaned surrounding inclination data and cleaned soil stress data; Performing spatiotemporal interpolation on the cleaned surface subsidence data to obtain a continuous surface subsidence field; Based on the surface settlement field, the abnormal settlement area is identified by using a curvature analysis algorithm to obtain a surface settlement state assessment result; Performing trend analysis on the cleaned surrounding inclination data to obtain a building inclination rate curve, and based on the building inclination rate curve, determining the stability level of surrounding buildings by a threshold discrimination method to obtain a surrounding inclination state assessment result; Performing stress path analysis on the cleaned soil stress data to obtain the stress evolution process of key parts, and based on the stress evolution process, calculating the soil stability index through the plastic zone expansion prediction model to obtain the soil stress state assessment result; The surface settlement state assessment results, the surrounding inclination state assessment results and the soil stress state assessment results are fused to obtain a real-time state parameter set.

5. The design and construction scheme optimization method for subway double-sided extension and heave control according to claim 1 is characterized in that: Based on the real-time status parameter set, the adjustable schedule corresponding to the subway station expansion construction process is updated to obtain an updated schedule and control measure data, wherein the control measure data includes active precipitation coupled deformation control measures and station top interval dynamic loading control measures, including: Performing a risk level assessment on the real-time status parameter set to obtain a current risk index for the subway station construction, and performing a deviation analysis on the adjustable schedule based on the current risk index to obtain a schedule deviation matrix; Performing decision analysis on the schedule deviation matrix to obtain a schedule adjustment strategy set, and optimizing time parameters of the adjustable schedule based on the schedule adjustment strategy set to obtain a preliminary updated schedule; Re-allocating resources for the preliminary update schedule to obtain an update schedule after resource balancing, and optimizing parameters of the active dewatering system based on the update schedule to obtain a dewatering well layout plan and a pumping rate curve; A coupling analysis is performed on the dewatering well arrangement scheme and the pumping rate curve to obtain active dewatering coupled deformation control measures, and based on the update schedule, a loading simulation is performed on the station top interval to obtain a dynamic loading force distribution diagram; The dynamic loading force distribution diagram is time-series optimized to obtain the station top interval dynamic loading control measures, and the active precipitation coupled deformation control measures and the station top interval dynamic loading control measures are integrated into a control measure data package.

6. The design and construction scheme optimization system for subway double-sided expansion and sinking control is characterized by: The design and construction scheme optimization system for subway double-sided extension and construction heave and sink control includes: A processing module is used to collect and preprocess multi-source data related to the subway station during the bilateral expansion construction process to obtain a standardized data set, wherein the standardized data set includes design parameters, monitoring data, and construction management data; A fusion module is used to perform multi-dimensional fusion analysis on the standardized data set to obtain target evaluation data, wherein the target evaluation data includes: earth excavation evaluation data, support system evaluation data and deformation control evaluation data; An optimization module is used to perform intelligent optimization processing on the preset construction plan based on the target evaluation data to obtain an optimized construction plan, wherein the optimized construction plan includes a layered and block excavation construction sub-plan, a counter-pressure island strip drawing construction sub-plan, and an early closure construction sub-plan; It is specifically used to perform multi-objective decomposition on the target evaluation data, obtain weight coefficients of earth excavation targets, support system targets and deformation control targets, and based on the weight coefficients, preliminarily score the preset construction plans to obtain an initial scoring matrix; optimize and iterate the initial scoring matrix through a genetic algorithm to obtain a set of candidate plans that meet multi-objective requirements, and perform parameterized modeling on the layered and block excavation plans in the candidate plan set to obtain a layered and block model that reflects the excavation depth, block size and excavation sequence; based on the layered and block model, perform geological stress field simulation to obtain stress distribution data during the layered and block excavation process, and optimize the layered and block excavation plan through the stress distribution data to obtain the layered and block excavation construction sub-plan; perform geometric optimization on the back pressure island stripping plan in the candidate plan set , obtain the optimal back-pressure island shape and strip width parameters, and based on the optimal back-pressure island shape and strip width parameters, perform stability analysis to obtain the safety factor curve of the back-pressure island strip construction; optimize the back-pressure island strip scheme through the safety factor curve to obtain the back-pressure island strip construction sub-scheme; optimize the early closure scheme in the candidate scheme set to obtain the optimal closure time and closure sequence, and based on the optimal closure time and closure sequence, perform deformation prediction analysis to obtain the deformation control effect evaluation data of the early closure construction; optimize the early closure scheme through the deformation control effect evaluation data to obtain the early closure construction sub-scheme; perform data fusion on the layered and block excavation construction sub-scheme, the back-pressure island strip construction sub-scheme and the early closure construction sub-scheme to obtain the optimized construction scheme; An analysis module is used to analyze the construction progress of the optimized construction plan based on a preset special process database through a critical path algorithm to obtain an adjustable progress plan, wherein the special process database includes: a reserved column process and a reverse loading process; The reserved column process in the special process database is parameterized to obtain a key parameter set for the reserved column construction, and based on the key parameter set, the reserved column construction is decomposed to obtain a detailed process network for the reserved column construction; the reverse loading process in the special process database is time-series analyzed to obtain the optimal implementation time of the reverse loading, and based on the optimal implementation time, the resource allocation of the reverse loading process is optimized to obtain a resource demand curve for the reverse loading; based on the resource demand curve for the reverse loading, the water injection process of the prefabricated steel blocks in the reverse loading process is process-analyzed to obtain the quality control points of the water injection of the prefabricated steel blocks, and based on the quality control points, the time of the water injection process of the prefabricated steel blocks is estimated to obtain the duration distribution of the water injection process; the sub-schemes of layered and block excavation, the sub-scheme of strip drawing of the counter-pressure island and the sub-scheme of early closure in the optimized construction plan are integrated to obtain a comprehensive construction network diagram; the comprehensive construction network diagram is calculated by the critical path algorithm to obtain the adjustable schedule; A monitoring module, for performing real-time monitoring and analysis on the construction process of the subway station based on the adjustable progress plan, and obtaining a real-time state parameter set, wherein the real-time state parameter set includes: surface settlement state, surrounding inclination state, and soil stress state; An updating module is used to update the adjustable progress plan corresponding to the subway station expansion construction process based on the real-time status parameter set, and obtain an updated progress plan and control measures data, wherein the control measures data include active precipitation coupled deformation control measures and station top interval dynamic loading control measures.

Images