A decision-making system and analysis method for rail transit engineering data analysis

Abstract

The present invention proposes a decision-making system and analysis method for rail transit engineering data analysis, the method comprising: constructing a first virtual digital twin model according to first data related to the rail transit engineering; analyzing the first data to obtain first construction risk data, first key node data and first countermeasures; obtaining multiple first construction plans according to the first virtual digital twin model, the first construction risk data, the first key node data and the first countermeasures; obtaining a first construction model by combining the first data and the first construction plan; installing multiple IoT smart sensors for collecting second data according to the first construction model; developing a construction decision support system based on a rule engine and artificial intelligence according to the first construction model and the second data, and providing decision suggestions. The scheme of the present invention can not only intelligently make flexible construction management decisions based on engineering data, but also reduce manual participation, and improve the efficiency and accuracy of construction management decisions.

Description

Technical Field

[0001] The present invention relates to the field of rail transit technology, and in particular to a decision-making system and an analysis method for rail transit engineering data analysis. Background Art

[0002] Most of the existing decision-making systems for rail transit engineering data analysis rely on simple statistical analysis models, and do not use more complex AI technology for in-depth analysis. The output analysis results generally cannot explain the analysis process well, which is not conducive to the understanding and application of engineers. Some systems also require a large number of manual participation in the analysis and decision-making process, and the degree of automation needs to be improved. An intelligent, efficient and accurate decision-making solution for rail transit engineering data analysis is urgently needed. Summary of the invention

[0003] Based on the above problems, the present invention proposes a decision-making system and analysis method for rail transit engineering data analysis. Through the scheme of the present invention, using artificial intelligence and big data analysis technology, not only can flexible construction management decisions be made intelligently based on engineering data, but also manual participation can be reduced, thereby improving the efficiency and accuracy of construction management decisions.

[0004] In view of this, one aspect of the present invention proposes a decision-making system for rail transit engineering data analysis, including: a cloud server, an IoT smart sensor and an IoT server; wherein,

[0005] The cloud server is configured as follows:

[0006] Acquiring first data related to the rail transit project;

[0007] Constructing a first virtual digital twin model including all rail transit engineering data according to the first data;

[0008] Applying big data technology, summarizing and analyzing the first data, obtaining first construction risk data and first key node data, and formulating corresponding first response measures;

[0009] Performing scheme simulation and collision detection according to the first virtual digital twin model, the first construction risk data, the first key node data, and the first countermeasures to obtain a plurality of first construction schemes;

[0010] Combining the first data and the first construction plan and using a deep learning algorithm to train a first construction model including a construction guidance model, a construction progress prediction model, a quality risk model, a safety accident risk model, and a preventive measures planning model;

[0011] According to the first construction model, a plurality of the IoT smart sensors are installed at the construction site to collect second data during the construction process;

[0012] A knowledge graph and a decision tree are established based on the first construction model and the second data, and a construction decision support system based on a rule engine and artificial intelligence is developed to provide decision recommendations.

[0013] Another aspect of the present invention provides a decision analysis method for rail transit engineering data analysis, comprising:

[0014] Acquiring first data related to the rail transit project;

[0015] Constructing a first virtual digital twin model including all rail transit engineering data according to the first data;

[0016] Applying big data technology, summarizing and analyzing the first data, obtaining first construction risk data and first key node data, and formulating corresponding first response measures;

[0017] Performing scheme simulation and collision detection according to the first virtual digital twin model, the first construction risk data, the first key node data, and the first countermeasures to obtain a plurality of first construction schemes;

[0018] Combining the first data and the first construction plan and using a deep learning algorithm to train a first construction model including a construction guidance model, a construction progress prediction model, a quality risk model, a safety accident risk model, and a preventive measures planning model;

[0019] According to the first construction model, a plurality of IoT smart sensors are installed at the construction site to collect second data during the construction process;

[0020] A knowledge graph and a decision tree are established based on the first construction model and the second data, and a construction decision support system based on a rule engine and artificial intelligence is developed to provide decision recommendations.

[0021] Optionally, the step of applying big data technology to summarize and analyze the first data to obtain first construction risk data and first key node data, and formulating corresponding first response measures includes:

[0022] Cleaning, denoising, and integrating the first data to form a structured first data set, and using the first data set as a basic data source for big data analysis;

[0023] Using statistics and data mining techniques, analyzing the first data set to identify key influencing factors and establishing a first model reflecting project progress and risks;

[0024] Based on the first model, a risk identification algorithm is developed to dynamically identify potential problems and risks according to the project schedule and quality requirements, thereby obtaining the first construction risk data;

[0025] In combination with the project schedule, identify the key nodes with tight construction period and resources and the important process nodes that have the greatest impact on the overall project schedule, and obtain the first key node data from the key nodes and the important process nodes;

[0026] According to the first construction risk data and the first key node data, using the expert knowledge base, a corresponding emergency plan and resource scheduling solution are given as the first response measure.

[0027] Optionally, the step of using statistics and data mining techniques to analyze the first data set to identify key influencing factors and establish a first model reflecting project progress and risks includes:

[0028] extracting a first characteristic variable related to project progress and risk from the first data set through domain knowledge;

[0029] Using statistical methods to identify the correlation between the first characteristic variables, and deleting redundant variables to obtain second characteristic variables;

[0030] Using a machine learning method to analyze the importance of each of the second characteristic variables to determine the key influencing factors;

[0031] Combining the second characteristic variable and the key influencing factor, using a regression algorithm to train a first project schedule prediction model;

[0032] Combining the second characteristic variable and the key influencing factor, using a classification algorithm to train a first risk classification model;

[0033] The first project schedule prediction model and the first risk classification model are integrated into the first model.

[0034] Optionally, the step of developing a risk identification algorithm based on the first model, dynamically identifying potential problems and risks according to the project schedule and quality requirements, and obtaining the first construction risk data includes:

[0035] Collect historical project risk cases and build a standardized first risk database;

[0036] Acquire real-time progress and real-time quality data of the project, and match the first risk type that is expected to occur according to the trigger conditions in the first risk library;

[0037] Using the first risk classification model, evaluating the matched first risk type to obtain a first risk evaluation result;

[0038] Using the first project progress prediction model, predict the subsequent progress and completion time of the project to obtain a first progress prediction result;

[0039] combining the first risk assessment result with the first progress prediction result to dynamically identify the first construction risk data that may affect the project progress;

[0040] The identified first construction risk data is saved in the form of a structured data set.

[0041] Optionally, the step of identifying, in combination with the project schedule, key nodes where construction period and resources are tight and important process nodes that have the greatest impact on the overall project schedule, and acquiring the first key node data from the key nodes and the important process nodes includes:

[0042] Extracting a pipe jacking construction schedule from the project schedule;

[0043] The data of key points in the pipe jacking construction schedule, including geological condition data, tunnel burial depth data, tunnel spacing, and starting and arrival engineering data of the pipe jacking, are identified as the first key node data.

[0044] Optionally, the step of providing a corresponding emergency plan and resource scheduling solution as the first response measure using an expert knowledge base according to the first construction risk data and the first key node data includes:

[0045] Collect emergency plan templates and resource scheduling plan examples in the construction field to form a structured emergency plan knowledge base;

[0046] Determining a matching first emergency plan template in the emergency plan knowledge base according to the first construction risk data;

[0047] Determine a matching first resource scheduling scheme according to the first key node data and resource status data;

[0048] Evaluate the first emergency plan template and the first resource scheduling scheme, and adjust and improve the first emergency plan template and the first resource scheduling scheme according to the evaluation results;

[0049] Simulating the execution effect of the first emergency plan template and the first resource scheduling solution in the first virtual digital twin system;

[0050] The first emergency plan template and the first resource scheduling plan are adjusted according to the execution effect to obtain the first response measure.

[0051] Optionally, the step of performing scheme simulation and collision detection according to the first virtual digital twin model, the first construction risk data, the first key node data and the first countermeasures to obtain a plurality of first construction schemes includes:

[0052] Importing the first construction risk data and the first key node data into the first virtual digital twin model for association and annotation;

[0053] Setting simulation parameters of resources, environment, and construction process in the first virtual digital twin model, and defining simulation conditions of different construction schemes;

[0054] Simulate the execution of different construction plans and calculate the time schedule, resource utilization rate and risk elimination effect of each construction plan;

[0055] During the simulation, a collision detection algorithm is used to check whether different construction plans conflict in time and space;

[0056] Based on the simulation and collision detection results, multiple feasible construction plans that meet the time, resource and risk constraints are generated, and quantitative and qualitative evaluation and analysis are performed on different feasible construction plans to determine multiple first construction plans.

[0057] Optionally, the step of installing a plurality of IoT smart sensors at the construction site according to the first construction model for collecting second data during the construction process includes:

[0058] Determining the data input requirements of the model for sensor data according to the first construction model;

[0059] Determine the types of sensors to be installed and the sensor coverage of various sensors according to the data input requirements;

[0060] Calculating the required number of sensors according to the construction site area, distribution of important areas, and coverage of the sensors;

[0061] Formulate a first distribution plan of sensors according to the data requirements of different regions of the first construction model, the sensor types, and the number of sensors;

[0062] Obtaining power supply data, network wiring data, and network coverage data of the construction area;

[0063] According to the power supply data, the network wiring data, and the network coverage data, the first virtual digital twin model is used to simulate, verify, and modify the first distribution plan to ensure that the data input requirements of the first construction model can be met;

[0064] Install a plurality of the IoT smart sensors according to the first distribution scheme to collect the second data;

[0065] During the construction process, the sensor layout is dynamically adjusted according to changes in the data input requirements of the first construction model.

[0066] Optionally, the step of combining the first data and the first construction plan and using a deep learning algorithm to train a first construction model including a construction guidance model, a predicted construction progress model, a quality risk model, a safety accident risk model, and a preventive measures planning model comprises:

[0067] Extracting first construction data including BIM data, sensor data, document data, video data, and audio data from the first data;

[0068] Extracting data about construction instructions from the first construction data, and building the construction instruction model based on the CNN neural network in combination with the first construction plan;

[0069] Extracting data on project progress and resource utilization from the first construction data, and building the LSTM-based predicted construction progress model in combination with the first construction plan;

[0070] Extracting usage quality inspection data and quality requirement data from the first construction data, and building the quality risk model based on a convolutional neural network in combination with the first construction plan;

[0071] Extracting safety record data and safety specification data from the first construction data, and training the multi-modal safety accident risk model in combination with the first construction plan;

[0072] Extracting historical preventive measures data and preset preventive requirements data from the first construction data, and building the preventive measures planning model based on the knowledge graph in combination with the first construction plan;

[0073] The first construction model driven by deep learning is constructed by integrating the construction guidance model, the predicted construction progress model, the quality risk model, the safety accident risk model and the preventive measures planning model.

[0074] According to the technical solution of the present invention, the decision analysis method for rail transit engineering data analysis includes: obtaining first data related to rail transit engineering; constructing a first virtual digital twin model containing all rail transit engineering data according to the first data; applying big data technology to summarize and analyze the first data, obtain first construction risk data, first key node data, and formulate corresponding first countermeasures; according to the first virtual digital twin model, the first construction risk data, the first key node data and the first countermeasures, perform scheme simulation and collision detection to obtain multiple first construction schemes; combining the first data and the first construction scheme and using deep learning algorithm training to obtain a first construction model including a construction guidance model, a predicted construction progress model, a quality risk model, a safety accident risk model and a preventive measures planning model; according to the first construction model, multiple IoT smart sensors are installed at the construction site for collecting second data during the construction process; according to the first construction model and the second data, a knowledge map and a decision tree are established, and a construction decision support system based on a rule engine and artificial intelligence is developed to provide decision suggestions. Through the scheme of the present invention, using artificial intelligence and big data analysis technology, not only can flexible construction management decisions be made intelligently based on engineering data, but also human participation is reduced, and the efficiency and accuracy of construction management decisions are improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0075] Figure 1 is a schematic block diagram of a decision system for rail transit engineering data analysis provided by an embodiment of the present invention;

[0076] Figure 2 It is a flow chart of a decision analysis method for rail transit engineering data analysis provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0077] In order to more clearly understand the above-mentioned purpose, features and advantages of the present invention, the present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be noted that the embodiments of the present application and the features in the embodiments can be combined with each other without conflict.

[0078] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Therefore, the protection scope of the present invention is not limited to the specific embodiments disclosed below.

[0079] The terms "first", "second", etc. in the specification and claims of this application and the above-mentioned drawings are used to distinguish different objects, rather than to describe a specific order. In addition, the terms "including" and "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 limited to the listed steps or units, but optionally includes steps or units that are not listed, or optionally includes other steps or units inherent to these processes, methods, products or devices.

[0080] Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present application. The appearance of the phrase in various locations in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0081] Refer to the following Figure 1 to Figure 2 A decision system and analysis method for rail transit engineering data analysis provided according to some embodiments of the present invention will be described.

[0082] like Figure 1 As shown, one embodiment of the present invention provides a decision system for rail transit engineering data analysis, including:

[0083] Cloud servers, IoT smart sensors and IoT servers; among them,

[0084] The cloud server is configured as follows:

[0085] Acquiring first data related to the rail transit project;

[0086] Constructing a first virtual digital twin model containing all rail transit project data based on the first data (the model can reflect the overall situation and details of the project in real time and assist decision-making);

[0087] Applying big data technology, summarizing and analyzing the first data, obtaining first construction risk data and first key node data, and formulating corresponding first response measures;

[0088] Performing scheme simulation and collision detection according to the first virtual digital twin model, the first construction risk data, the first key node data, and the first countermeasures to obtain a plurality of first construction schemes;

[0089] Combining the first data and the first construction plan and using a deep learning algorithm to train a first construction model including a construction guidance model, a construction progress prediction model, a quality risk model, a safety accident risk model, and a preventive measures planning model;

[0090] According to the first construction model, a plurality of the IoT smart sensors are installed at the construction site to collect second data during the construction process;

[0091] A knowledge graph and a decision tree are established based on the first construction model and the second data, and a construction decision support system based on a rule engine and artificial intelligence is developed to provide decision-making recommendations (for example, using computer vision technology for image recognition and analysis to monitor construction progress, quality and safety, and to assist in decision-making).

[0092] In some possible implementations of the present invention, the step of establishing a knowledge graph and a decision tree according to the first construction model and the second data, developing a construction decision support system based on a rule engine and artificial intelligence, and providing decision suggestions includes:

[0093] Extracting construction knowledge data in terms of construction plan, resources, risks, and quality management from the second data;

[0094] Using knowledge graph technology, establish the relationship between the conceptual entities involved in the construction knowledge data to form a construction field knowledge graph;

[0095] Using a decision tree algorithm, learning the first construction model and the construction knowledge data, and obtaining a construction field decision tree reflecting the decision logic;

[0096] Developing a reasoning engine based on predefined rules according to the construction field knowledge graph and the construction field decision tree;

[0097] Integrate the deep learning algorithm trained in the first construction model to introduce data-driven decision-making capabilities;

[0098] Use NLP technology to develop text or voice interactive interfaces to receive decision-making queries;

[0099] Use visualization technology to present the decision-making process in a visual flowchart or experiential interface;

[0100] Use new decision cases to continuously optimize knowledge graphs, decision trees, and deep learning models.

[0101] In this embodiment, through the above process, an intelligent and efficient decision support system can be finally constructed.

[0102] Some possible implementations of the present invention also include: using VR equipment to allow decision makers to view design plans and construction simulations in an immersive way, better understand project conditions, and make decisions; establishing an information-based collaborative work platform to allow all parties involved in the project to share project data and decision-making information in real time to improve decision-making efficiency; carrying out predictive maintenance and accident warnings, using machine learning to predict equipment failures and accidents, achieving predictive maintenance and accident warnings, and taking precautions in advance.

[0103] In the embodiment of the present invention, through sufficient data collection, analysis and application of intelligent technology, a scientific construction plan for the rail transit project can be formulated to ensure the quality and progress of the project, reduce safety accidents and lower construction risks, which is particularly critical for complex urban rail transit projects.

[0104] In this embodiment, the first data related to the rail transit project includes, but is not limited to: line design drawings, surrounding building (structure) survey data, road survey data, bridge survey data, tunnel survey data, river survey data, cultural relics survey data, pipeline survey data, geological exploration report, surrounding environment assessment report, construction area and surrounding area within a preset distance three-dimensional point cloud data and image data, etc., these data can reflect the specific situation of the project. For example, the whole line of the line is geologically surveyed, and data such as soil layer conditions, underground hydrogeological conditions, and environmental protection conditions are collected. These data can analyze the difficulties and risks of the project; collect data on the impact assessment of the surrounding environment, including noise, vibration, traffic flow, etc., which can evaluate the impact of the construction on the surrounding environment; collect municipal pipeline information and determine the configuration to prevent the construction from causing damage to municipal facilities; conduct special research on key sections and major bridges and tunnels, collect historical data, and determine a suitable construction plan.

[0105] The technical solution of this embodiment is adopted, by obtaining the first data related to the rail transit project; constructing the first virtual digital twin model containing all rail transit project data according to the first data; applying big data technology, summarizing and analyzing the first data, obtaining the first construction risk data, the first key node data, and formulating the corresponding first countermeasures; performing scheme simulation and collision detection according to the first virtual digital twin model, the first construction risk data, the first key node data and the first countermeasures, and obtaining multiple first construction schemes; combining the first data and the first construction scheme and using the deep learning algorithm to train to obtain the first construction model including the construction guidance model, the predicted construction progress model, the quality risk model, the safety accident risk model and the preventive measures planning model; according to the first construction model, installing multiple IoT smart sensors at the construction site for collecting the second data during the construction process; establishing the knowledge graph and decision tree according to the first construction model and the second data, developing the construction decision support system based on the rule engine and artificial intelligence, and providing decision suggestions. Through the scheme of the present invention, using artificial intelligence and big data analysis technology, not only can flexible construction management decisions be made intelligently according to engineering data, but also human participation is reduced, and the efficiency and accuracy of construction management decisions are improved.

[0106] It should be known that Figure 1 The block diagram of the decision-making system for rail transit engineering data analysis is shown for illustration only, and the number of modules shown therein does not limit the protection scope of the present invention.

[0107] See also Figure 2 Another embodiment of the present invention provides a decision analysis method for rail transit engineering data analysis, comprising:

[0108] Acquiring first data related to the rail transit project;

[0109] Based on the first data (using BIM, GIS and VR technology), a first virtual digital twin model containing all rail transit project data is constructed (the model can reflect the overall situation and details of the project in real time and assist decision-making);

[0110] Applying big data technology, summarizing and analyzing the first data, obtaining first construction risk data and first key node data, and formulating corresponding first response measures;

[0111] Performing scheme simulation and collision detection according to the first virtual digital twin model, the first construction risk data, the first key node data, and the first countermeasures to obtain a plurality of first construction schemes;

[0112] Combining the first data and the first construction plan and using a deep learning algorithm to train a first construction model including a construction guidance model, a construction progress prediction model, a quality risk model, a safety accident risk model, and a preventive measures planning model;

[0113] According to the first construction model, a plurality of IoT smart sensors are installed at the construction site to collect second data during the construction process;

[0114] A knowledge graph and a decision tree are established based on the first construction model and the second data, and a construction decision support system based on a rule engine and artificial intelligence is developed to provide decision-making recommendations (for example, using computer vision technology for image recognition and analysis to monitor construction progress, quality and safety, and to assist in decision-making).

[0115] In some possible implementations of the present invention, the step of establishing a knowledge graph and a decision tree according to the first construction model and the second data, developing a construction decision support system based on a rule engine and artificial intelligence, and providing decision suggestions includes:

[0116] Extracting construction knowledge data in terms of construction plan, resources, risks, and quality management from the second data;

[0117] Using knowledge graph technology, establish the relationship between the conceptual entities involved in the construction knowledge data to form a construction field knowledge graph;

[0118] Using a decision tree algorithm, learning the first construction model and the construction knowledge data, and obtaining a construction field decision tree reflecting the decision logic;

[0119] Developing a reasoning engine based on predefined rules according to the construction field knowledge graph and the construction field decision tree;

[0120] Integrate the deep learning algorithm trained in the first construction model to introduce data-driven decision-making capabilities;

[0121] Use NLP technology to develop text or voice interactive interfaces to receive decision-making queries;

[0122] Use visualization technology to present the decision-making process in a visual flowchart or experiential interface;

[0123] Use new decision cases to continuously optimize knowledge graphs, decision trees, and deep learning models.

[0124] In this embodiment, through the above process, an intelligent and efficient decision support system can be finally constructed.

[0125] In this embodiment, the specific functions of the rule engine are:

[0126] Encoding expert knowledge: The rule engine can encode the construction decision-making knowledge and experience provided by experts into several clear rules, such as "if there is heavy rain, stop working at heights."

[0127] Rule integration: Integrate the encoded rules into the rule engine to form a rule base containing a variety of decision-making knowledge.

[0128] Perform rule reasoning: In a given specific situation, the rule engine can automatically perform multi-rule joint reasoning based on the rule base to come up with decision recommendations.

[0129] Rule explanation: The rule engine can explain the process of generating decision recommendations and explain which rules were used.

[0130] Rule management: Add and adjust new rules, and manage rule execution priorities.

[0131] Case-based evolution: Continuously optimize and add rules through new decision-making cases to achieve rule evolution.

[0132] Combined with AI models: Fusion with deep learning models makes decisions both explainable and data-driven.

[0133] Therefore, the rule engine provides explainable reasoning and decision-making capabilities for decision support systems.

[0134] Some possible implementations of the present invention also include: using VR equipment to allow decision makers to view design plans and construction simulations in an immersive way, better understand project conditions, and make decisions; establishing an information-based collaborative work platform to allow all parties involved in the project to share project data and decision-making information in real time to improve decision-making efficiency; carrying out predictive maintenance and accident warnings, using machine learning to predict equipment failures and accidents, achieving predictive maintenance and accident warnings, and taking precautions in advance.

[0135] In the embodiment of the present invention, through sufficient data collection, analysis and application of intelligent technology, a scientific construction plan for the rail transit project can be formulated to ensure the quality and progress of the project, reduce safety accidents and lower construction risks, which is particularly critical for complex urban rail transit projects.

[0136] In this embodiment, the first data related to the rail transit project includes, but is not limited to: line design drawings, surrounding building (structure) survey data, road survey data, bridge survey data, tunnel survey data, river survey data, cultural relics survey data, pipeline survey data, geological exploration report, surrounding environment assessment report, construction area and surrounding area within a preset distance three-dimensional point cloud data and image data, etc., these data can reflect the specific situation of the project. For example, the whole line of the line is geologically surveyed, and data such as soil layer conditions, underground hydrogeological conditions, and environmental protection conditions are collected. These data can analyze the difficulties and risks of the project; collect data on the impact assessment of the surrounding environment, including noise, vibration, traffic flow, etc., which can evaluate the impact of the construction on the surrounding environment; collect municipal pipeline information and determine the configuration to prevent the construction from causing damage to municipal facilities; conduct special research on key sections and major bridges and tunnels, collect historical data, and determine a suitable construction plan.

[0137] In some possible implementations of the present invention, the step of applying big data technology to summarize and analyze the first data to obtain first construction risk data and first key node data, and formulating corresponding first response measures includes:

[0138] Cleaning, denoising, and integrating the first data to form a structured first data set, and using the first data set as a basic data source for big data analysis;

[0139] Using statistics and data mining techniques, analyzing the first data set to identify key influencing factors and establishing a first model reflecting project progress and risks;

[0140] Based on the first model, a risk identification algorithm is developed to dynamically identify potential problems and risks according to the project schedule and quality requirements, thereby obtaining the first construction risk data;

[0141] In combination with the project schedule, identify the key nodes with tight construction period and resources and the important process nodes that have the greatest impact on the overall project schedule, and obtain the first key node data from the key nodes and the important process nodes;

[0142] According to the first construction risk data and the first key node data, using the expert knowledge base, a corresponding emergency plan and resource scheduling solution are given as the first response measure.

[0143] In this embodiment, the risks, key node identification and response results can also be displayed through visualization means for decision makers to review, and feedback can be collected for model tuning; the model can be iterated using the newly added data to make risk identification and response more intelligent, forming a closed-loop feedback.

[0144] In some possible implementations of the present invention, in order to establish a smarter and more accurate model, the step of using statistics and data mining techniques to analyze the first data set to identify key influencing factors and establish a first model reflecting project progress and risks includes:

[0145] extracting a first characteristic variable related to project progress and risk from the first data set through domain knowledge;

[0146] Using statistical methods to identify the correlation between the first characteristic variables, and deleting redundant variables to obtain second characteristic variables;

[0147] Using a machine learning method to analyze the importance of each of the second characteristic variables to determine the key influencing factors;

[0148] Combining the second characteristic variable and the key influencing factor, using a regression algorithm to train a first project schedule prediction model;

[0149] Combining the second characteristic variable and the key influencing factor, using a classification algorithm to train a first risk classification model;

[0150] The first project schedule prediction model and the first risk classification model are integrated into the first model.

[0151] In this embodiment, it also includes: evaluating the indicators of the first model and selecting the best model; continuously optimizing the generalization ability of the model through parameter adjustment, ensemble learning and other methods; retraining the model with new data every week / month to adapt it to the new data distribution and realize dynamic updating; applying the trained model to real-time project progress prediction and risk identification.

[0152] In this embodiment, domain knowledge specifically refers to:

[0153] Expertise in the engineering field: including expertise in construction processes, project planning, resource management, cost control, etc. This knowledge can guide the determination of factors affecting progress and risks.

[0154] Construction experience: Experienced project managers and engineers have accumulated construction focus and experience in dealing with unusual situations. These can help determine the importance of variables.

[0155] Internal knowledge of the enterprise: data and records of the enterprise’s past projects. These data can be used to discover the characteristics of the enterprise’s projects and identify universal and special variables.

[0156] Industry knowledge graph: Integrate knowledge resources such as public knowledge graphs and ontology libraries within the industry to assist in variable selection.

[0157] Consult expert opinions: Organize expert meetings to consult and demonstrate the rationality of variables. Experts provide authoritative verification.

[0158] Continuous learning and iteration: Continuously accumulate knowledge and experience through new projects, iteratively improve the domain knowledge system, and enhance the quality of variable selection.

[0159] Comprehensively utilizing engineering domain knowledge obtained from various channels can effectively guide feature engineering and select high-quality variables, which is the key foundation of intelligent analysis.

[0160] In some possible implementations of the present invention, the step of developing a risk identification algorithm based on the first model, dynamically identifying potential problems and risks according to the project schedule and quality requirements, and obtaining the first construction risk data includes:

[0161] Collect historical project risk cases and build a standardized first risk database (including risk type, triggering conditions, impact level, etc.);

[0162] Acquire real-time progress and real-time quality data of the project, and match the first risk type that is expected to occur according to the trigger conditions in the first risk library;

[0163] Using the first risk classification model, evaluating the matched first risk type to obtain a first risk evaluation result;

[0164] Using the first project progress prediction model, predict the subsequent progress and completion time of the project to obtain a first progress prediction result;

[0165] combining the first risk assessment result with the first progress prediction result to dynamically identify the first construction risk data that may affect the project progress;

[0166] The identified first construction risk data is saved in the form of a structured data set (including risk type, possible impact, probability of occurrence, etc.).

[0167] Through the implementation plan, the accuracy and efficiency of obtaining construction risk data are further improved.

[0168] In some possible implementations of the present invention, the step of identifying the key nodes with tight construction period and resources and the important process nodes that have the greatest impact on the overall project schedule in combination with the project schedule, and acquiring the first key node data from the key nodes and the important process nodes includes:

[0169] Extracting a pipe jacking construction schedule from the project schedule;

[0170] The identification of the construction schedule of the pipe jacking includes geological condition data (in actual construction, the soil covering the pipe jacking tunnel mainly includes miscellaneous fill, silt, and silty clay. In particular, the silt stratum is highly sensitive and easily disturbed during the pipe jacking construction, resulting in uplift or settlement deformation of the surface), tunnel depth data (for example, the burial depth of the pipe jacking tunnel is 9.75m, the span of a single pipe jacking is 11.1m, and the span-cover ratio is only 0.89. The surrounding soil has a weak constraint on the soil covering the tunnel, and the surface is prone to deformation during construction), tunnel spacing (for example, the pipe jacking tunnel adopts a small spacing The data of the key points of the jacking construction are as follows: the distance between the two jacking tunnels is 1.5m, and the disturbance of the soil during the construction process may produce a superposition effect, causing the soil on the top of the tunnel to be disturbed multiple times), the starting and arrival engineering data of the jacking pipe (the starting and arrival of the jacking pipe are the most problematic links in the jacking construction. Inaccurate positioning and poor direction control during the starting and arrival of the jacking construction, or quality problems of the end reinforcement often lead to collapse, mud burst, and water gushing at the time of starting and arriving, seriously affecting the construction safety.

[0171] It is understandable that in the construction of rail transit projects, pipe jacking construction is an important link. In this embodiment, selecting the data of key points in pipe jacking construction as the first key node data as the basis for construction management decision-making can improve the efficiency of management decision-making.

[0172] In some possible implementations of the present invention, the step of providing a corresponding emergency plan and resource scheduling solution as the first response measure using an expert knowledge base based on the first construction risk data and the first key node data includes:

[0173] Collect emergency plan templates and resource scheduling plan examples in the construction field to form a structured emergency plan knowledge base;

[0174] Determining a matching first emergency plan template in the emergency plan knowledge base according to the first construction risk data;

[0175] Determine a matching first resource scheduling scheme according to the first key node data and resource status data;

[0176] Evaluate the first emergency plan template and the first resource scheduling scheme, and adjust and improve the first emergency plan template and the first resource scheduling scheme according to the evaluation results;

[0177] Simulating the execution effect of the first emergency plan template and the first resource scheduling solution in the first virtual digital twin system;

[0178] The first emergency plan template and the first resource scheduling plan are adjusted according to the execution effect to obtain the first response measure.

[0179] In this embodiment, by combining the knowledge base and historical experience wisdom, rapid and efficient generation of countermeasures is achieved.

[0180] In some possible implementations of the present invention, the step of performing scheme simulation and collision detection according to the first virtual digital twin model, the first construction risk data, the first key node data and the first countermeasures to obtain multiple first construction schemes includes:

[0181] Importing the first construction risk data and the first key node data into the first virtual digital twin model for association and annotation;

[0182] Setting simulation parameters of resources, environment, and construction process in the first virtual digital twin model, and defining simulation conditions of different construction schemes;

[0183] According to the simulation conditions, simulate and execute different construction plans, and calculate the time schedule, resource utilization rate, and risk elimination effect of each construction plan;

[0184] During the simulation, a collision detection algorithm is used to check whether different construction plans conflict in time and space;

[0185] Based on the simulation and collision detection results, multiple feasible construction plans that meet the time, resource and risk constraints are generated, and quantitative and qualitative evaluation and analysis are performed on different feasible construction plans to determine multiple first construction plans.

[0186] In this embodiment, simulation is performed by combining the first construction risk data, the first key node data, and the first virtual digital twin model to obtain a first construction plan that accurately meets the requirements.

[0187] In some possible implementations of the present invention, the step of installing a plurality of IoT smart sensors at the construction site according to the first construction model to collect second data during the construction process includes:

[0188] Determining the data input requirements of the model for the sensor data (such as data requirement type, data requirement format, and data requirement amount, etc.) according to the first construction model;

[0189] Determine the types of sensors that need to be installed (e.g., video cameras, RFID sensors, environmental sensors, etc.) and the sensor coverage of various sensors based on the data input requirements;

[0190] Calculating the required number of sensors according to the construction site area, distribution of important areas, and coverage of the sensors;

[0191] Formulate a first distribution plan of sensors according to the data requirements of different regions of the first construction model, the sensor types, and the number of sensors;

[0192] Obtaining power supply data, network wiring data, and network coverage data of the construction area;

[0193] According to the power supply data, the network wiring data, and the network coverage data, the first virtual digital twin model is used to simulate, verify, and modify the first distribution plan to ensure that the data input requirements of the first construction model can be met;

[0194] Install a plurality of the IoT smart sensors according to the first distribution scheme to collect the second data;

[0195] During the construction process, the sensor layout is dynamically adjusted according to changes in the data input requirements of the first construction model.

[0196] In this embodiment, sensor types include video cameras, vibration sensors, noise sensors, etc., and the sensors can be arranged according to the characteristics of different local areas within the construction area (such as important areas and areas with high safety risks, with richer sensor types and denser numbers). This not only provides comprehensive monitoring data, but also maximizes resource utilization and saves costs.

[0197] In some possible implementations of the present invention, the step of combining the first data and the first construction plan and using a deep learning algorithm to train to obtain a first construction model including a construction guidance model, a predicted construction progress model, a quality risk model, a safety accident risk model, and a preventive measures planning model includes:

[0198] Extracting first construction data including BIM data, sensor data, document data, video data, and audio data from the first data;

[0199] Extracting data about construction instructions from the first construction data, and building the construction instruction model based on the CNN neural network in combination with the first construction plan;

[0200] Extracting data on project progress and resource utilization from the first construction data, and building the LSTM-based predicted construction progress model in combination with the first construction plan;

[0201] Extracting usage quality inspection data and quality requirement data from the first construction data, and building the quality risk model based on a convolutional neural network in combination with the first construction plan;

[0202] Extracting safety record data and safety specification data from the first construction data, and training the multi-modal safety accident risk model in combination with the first construction plan;

[0203] Extracting historical preventive measures data and preset preventive requirements data from the first construction data, and building the preventive measures planning model based on the knowledge graph in combination with the first construction plan;

[0204] The first construction model driven by deep learning is constructed by integrating the construction guidance model, the predicted construction progress model, the quality risk model, the safety accident risk model and the preventive measures planning model.

[0205] In this embodiment, a construction model including multiple sub-models is constructed by utilizing artificial intelligence technology, thereby making the overall solution smarter, more accurate, and more efficient.

[0206] The following will introduce some solutions involved in actual construction scenarios (without any limitation to the technical solution of the present invention) to help understand the technical solution of the present invention.

[0207] In rail transit projects, under the condition of the same burial depth, due to the large span of the top of the pipe jacking equipment and the pipe section, and the lack of arching of the pipe jacking equipment and the pipe section, the arching effect of the overlying soil is greatly weakened, and the sensitivity to deformation is strong. In addition, as the cross-section of the pipe jacking machine excavation increases, there may be uneven improvement of the ballast soil, and the pressure of the mud and water tank at each point in the super-large cross-section rectangular mud and water tank may be different, which will have an adverse effect on the stability of the excavation surface, and the excavation surface may be partially over-excavated. Or the sealing is not good, the thixotropic mud leaks, and the grouting quality control is not good, resulting in excessive settlement, which seriously affects the pipeline and ground transportation. Therefore, how to ensure the settlement of the rectangular pipe jacking construction has become a heavy and difficult point in the project.

[0208] 1. Factors affecting surface deformation:

[0209] (1) Geological conditions: The soil covering the tunnel of this project is mainly miscellaneous fill, silt, and silty clay. In particular, the silt stratum is highly sensitive and easily disturbed during the tunnel construction, resulting in uplift or settlement deformation of the surface.

[0210] (2) Tunnel burial depth: The burial depth of the tunnel for this project is 9.75 m, the span of a single tunnel is 11.1 m, and the cover-span ratio is only 0.89. The surrounding soil has a weak constraint on the soil covering the tunnel, and the ground surface is prone to deformation during construction.

[0211] (3) Tunnel spacing: The jacking tunnel is constructed with a small spacing. The distance between two jacking tunnels is 1.5 m. The disturbance of the soil during the construction process may produce a superposition effect, causing the soil on the top of the tunnel to be disturbed multiple times.

[0212] Construction measures:

[0213] During the actual advancement of the pipe jacking, the grouting amount (thixotropic mud), grouting location and grouting pressure are adjusted after analysis and judgment based on the ground settlement monitoring data.

[0214] During construction, thixotropic mud is injected when necessary to effectively control the later settlement and carry out information-based dynamic construction management.

[0215] When the pipe jacking is advanced under the pipeline, the parameters accumulated at the beginning are used for advancement. At the same time, dedicated personnel are required to monitor the settlement of the strata around the pipeline day and night, and promptly monitor the deformation of the pipeline. The monitoring data is promptly and accurately fed back to the pipe jacking construction work surface, so that the central control room can make correct judgments based on the situation reflected on the ground, and guide the timely optimization and adjustment of the pipe jacking advancement parameters.

[0216] At the same time, according to the ground load conditions, the mud balance setting value is recalculated in time and adjusted according to the ground heave value to enable the jacking pipe to advance quickly and evenly, shorten the crossing time as much as possible, prevent over-excavation and under-excavation, reduce the disturbance to the soil, minimize the loss of formation, control the settlement within the minimum range, and meet the settlement requirements.

[0217] Before the construction, the pipe jacking machine needs to be fully inspected to ensure that there are no mechanical failures during the crossing process, and daily maintenance should be strengthened:

[0218] (1) Before pipe jacking construction, the end of the excavation surface should be reinforced to reduce the disturbance of the surrounding soil caused by pipe jacking construction.

[0219] (2) Strengthen construction management and pass the overlying pipeline at a uniform and continuous speed.

[0220] (3) Control the pressure in the mud silo, adjust the viscosity and specific gravity of the thixotropic mud, avoid instability of the tunnel face, strengthen slag discharge management, and appropriately reduce the amount of slag discharge.

[0221] (4) Ensure that the synchronous grouting system (thixotropic mud) works normally and increase the grouting volume appropriately.

[0222] (5) By setting mud injection holes at the top of the pipe section, when the pipe deformation exceeds its allowable value, a mud pump is used to inject mud for timely filling to reduce settlement.

[0223] (6) Anti-leakage measures: strictly control the quality of pipe joint assembly to ensure that the pipe joints are evenly stressed during the jacking process to prevent leakage caused by cracking of the joints. During the project construction, the quality control of pipe joint assembly was strengthened and traditional EPDM rubber waterstops were installed to improve the waterproof performance of the pipe joints.

[0224] (7) Pre-adjustment of the pipe jacking machine’s posture: As the cutter head and machine body are heavy, the pipe jacking machine is prone to “lowering its head” during the initial operation. During construction, the initial posture is adjusted so that the machine body can “raise its head” appropriately. During the excavation process, the pipe jacking machine’s posture is promptly corrected and rectified. The principle of “small amount and frequent correction” is followed to avoid excessive correction and minimize the number of disturbances to the surrounding soil.

[0225] (8) Strengthening construction organization: During the construction of the pipe jacking tunnel, strengthen the construction organization, arrange the construction plan reasonably, complete the work content on time, ensure the supply of construction materials, ensure the slag discharge progress, and do a good job of external coordination to eliminate hidden dangers that may cause the project to stop, maintain the continuity of the tunnel jacking construction, and reduce the downtime of the pipe jacking machine.

[0226] The starting and arrival of pipe jacking are the links in the pipe jacking construction that are most prone to problems. Inaccurate positioning, poor direction control, or quality problems in the end reinforcement during the starting and arrival of pipe jacking construction often lead to collapse, mud bursts, and water gushing during the starting and arrival, seriously affecting construction safety.

[0227] 2. In addition, due to the limitation of conditions, the starting point of the jacking pipe needs to be set in the main structure section of the station, and there is no ready-made rear wall as a jacking force reaction device; and the cross-sectional size of the jacking pipe is large, and the jacking force provided by the reaction device is large, and the requirements for the rear wall are high; the single-hole jacking distance of this project is about 52.5m, and the standard value of the total jacking force reaches nearly 3,300 tons according to relevant calculations.

[0228] Construction measures:

[0229] (1) Strictly reinforce the end caps according to the length and width required by the design, and do a good job of process control; during the reinforcement construction process, strictly follow the relevant requirements, strictly control the construction quality from the perspectives of material delivery, equipment, construction technology, etc., to ensure the quality of reinforcement.

[0230] (2) After the end stratum reinforcement is completed, vertical (or horizontal) coring is carried out on the reinforced area to check the integrity and strength of the reinforcement.

[0231] (3) The initial reaction wall of the construction adopts the existing underground continuous wall + reaction wall and middle plate as the advancing reaction wall.

[0232] (4) Ensure waterproof sealing of the tunnel opening. When the pipe is pushed forward, install the tunnel door ring with embedded steel ring in advance. When starting, use a rubber sheet as the tunnel door seal, and reserve a grouting port above the tunnel door to avoid the pipe section rubbing against the tunnel door seal for a long time during the starting process and the entire advancement process, which may cause the seal to be damaged, thus ensuring the safety of the project.

[0233] (5) A wire brush is added to the starting tunnel gate ring to increase the sealing of the tunnel gate and maintain dynamic waterproofing of the tunnel gate.

[0234] (6) When the jacking pipe arrives, the reaction force of the jacking pipe face is destroyed, reduced, or lost, and the thixotropic mud is lost. As with the initial stage, the quality of the reinforcement of the portal end is the core.

[0235] (7) Control the posture of the jacking pipe, and complete the starting and arrival of the jacking pipe as quickly as possible while ensuring the normal amount of mucking. At the same time, take into full consideration the changes in the nature of the strata due to the reinforcement of the end strata, pay close attention to the sudden changes in the posture of the jacking pipe during advancement, measure and correct the deviation frequently, and inject high-viscosity mud behind the jacking machine when jacking the last three pipe sections to prevent leakage when arriving.

[0236] (8) Conduct drills on the departure and arrival safety emergency plan and reserve emergency locations before departure and arrival to ensure that manpower and materials are promptly deployed for emergency treatment in the event of an abnormal situation to ensure the safety of the project.

[0237] (9) The rubber water-stop ring is installed on the working shaft portal according to the design drawings. Since the rubber ring of the pipe jacking tunnel is easily worn by sand particles and loses its water-stopping function, a shield tail brush is added to the steel ring of the portal inside the water-stop ring. The shield tail brush can effectively prevent sand particles from flowing toward the rubber ring at the tunnel entrance, thereby ensuring the integrity of the tunnel entrance water-stop system.

[0238] (10) On the basis of the designed water-stopping scheme, two additional measures are added: one is to add 18 DN12.7 mm chemical grouting holes at the end of the pipe joint to be used for water-stopping by pressing chemical grout on the outside of the joint after the jacking is completed; the other is to add a ring of rubber water-stopping strip in the middle of the joint pad as the last active water-stopping measure.

[0239] III. Risk Identification and Classification

[0240] 1. Risk engineering category: engineering inherent risk;

[0241] Risk sources: departure and arrival of main pipe jacking;

[0242] Description of basic risk conditions: The jacking pipe is an extra-large section jacking pipe with an outer cross-sectional size of 11.1x8.1m. Inaccurate positioning and poor direction control during the start and arrival of construction, or quality problems with the end reinforcement, lead to collapse, mud bursts, and water gushing at the start and arrival. When the SMW construction piles at the end are pulled out when the steel sections are pulled out when the jacking pipe exits and enters the hole, spraying occurs due to improper control of the slurry pressure pushed by the jacking face, which seriously affects construction safety.

[0243] Risk classification: Level II;

[0244] Countermeasures: Reinforce the end in strict accordance with the required length and width, and do a good job of process control; during the reinforcement construction process, strictly follow the relevant requirements to ensure the quality of the end reinforcement. Before the tunnel portal is chiseled out, horizontal exploration holes are made at the corners and center positions within the tunnel portal to be chiseled out. The number of exploration holes is flexibly arranged according to the site. If the seepage is found, secondary grouting reinforcement is carried out in time to ensure the integrity and continuity of the reinforcement body, without reinforcement blind spots, and no sand flow and mud gushing at the start and arrival. Make a good waterproof seal at the tunnel entrance, pre-install the tunnel portal ring with embedded steel rings at the start of the jacking pipe, and use the curtain rubber plate road as the tunnel portal seal. Control the posture of the jacking pipe, and complete the start and arrival of the jacking pipe as quickly as possible while ensuring the normal amount of mucking. At the same time, fully considering the changes in the nature of the stratum due to the reinforcement of the end stratum, pay close attention to the sudden change of the jacking pipe posture during advancement, measure frequently, correct deviation frequently, and inject high-viscosity mud behind the jacking machine when jacking the last three pipe sections to prevent leakage at the arrival. When pulling out the steel of the SMW pile during entry and exit of the hole, the slurry pressure on the face of the jacking machine head should be controlled. As a key working condition, the deformation of the ground and pipeline should be closely monitored, and backfill measures and materials should be prepared in time.

[0245] Risk level after disposal: Level III.

[0246] 2. Risk engineering category: engineering inherent risk;

[0247] Risk source: close fitting construction of main pipe jacking;

[0248] Description of basic risk conditions: The main pipe jacking project consists of two parallel pipe jacking channels with a clear distance of 1.5m between the channels. The following problems are likely to occur during construction: (1) The lateral pressure generated during the jacking process is likely to affect the adjacent pipe jacking channels that have been implemented first, causing deformation and displacement of adjacent pipe sections, and even causing damage. (2) The soil around the formed channel is disturbed again by the adjacent pipe jacking construction, which is likely to cause the ground and pipeline settlement to overlap, causing damage to the surrounding environment.

[0249] Risk classification: Level I;

[0250] Countermeasures: Conduct detailed geological surveys and pipeline surveys before construction, formulate feasible construction plans and organize experts to review them, make technical and construction preparations in advance, and adjust the pipe jacking excavation parameters and grouting parameters in time according to the test data and surface settlement during the excavation process to ensure that the deformation of the adjacent pipe jacking structure is controlled within the design and specification requirements. During the excavation process, the jacking posture is controlled and adjusted through the axis deviation control technology and axis roll control technology to ensure the safety of the jacking close construction. During the jacking excavation process, synchronous grouting and secondary grouting should be done well. Grouting must follow the principle of "pressing first and then jacking, pressing as the jacking progresses, and timely grouting". The quality of grouting construction should be controlled by the principle of pressure control as the main and grouting volume control as the auxiliary. At the same time, during the excavation process, technical parameters such as slag improvement, mud and water bin pressure and slag discharge volume must be strictly controlled to prevent the surface settlement from exceeding the limit. Measures to strengthen the longitudinal connection of pipe segments: Since the jacking pipe section intersects with the three large pipes on the upper part in the plane, and the jacking pipe channel implemented later may produce lateral unbalanced thrust on the jacking pipe that has been completed, the longitudinal stiffness of each jacking pipe channel is strengthened, and post-tensioned prestressed steel strands are used to tie the pipe segments together longitudinally.

[0251] Risk level after disposal: Level III.

[0252] 3. Risk engineering category: engineering inherent risk;

[0253] Risk source: Construction of main pipe jacking cross channel

[0254] Description of basic risk conditions: The length of the jacking pipe cross channel is 1.5m. During the construction of breaking the pipe segment structure connection, there are large vibrations, quality problems with the stratum reinforcement, untimely process connection, and untimely installation of the initial steel support, which lead to collapse, mud bursts, and water gushing in the breaking range, seriously affecting construction safety, and causing settlement and deformation of the upper pipeline, resulting in damage.

[0255] Risk classification: Level II;

[0256] Countermeasures: The hole should be opened after the two jacking pipes are connected, and the sleeve valve pipe of the jacking pipe channel is grouted and the strength is reached, so as to prevent the deformation of the channel caused by grouting reinforcement and consolidation from affecting the waterproofing of the hole. Before opening the hole, the reserved grouting pipe should be used to grout the 1.5m stratum between the channels, and then reliable temporary support should be carried out inside the pipe sections without holes and the pipe sections within 3m on both sides. Then, the reserved holes should be broken in blocks, and the steel primary support should be erected, and the concrete should be sprayed and sealed in time. If the water consumption is found to be too large during this period, the surrounding reserved holes should be supplemented with grouting, or holes should be drilled and grouted again at nearby locations. After the entire primary support is closed, the secondary lining structure should be constructed in time.

[0257] Risk level after disposal: Level III.

[0258] 4. Risk engineering categories: environmental risks;

[0259] Risk sources: Guangzhou gas pipeline, Sinopec oil pipeline, Dapeng gas pipeline;

[0260] Description of basic risk conditions: The jacking pipe intersects the pipeline, and the jacking pipe construction passes under the pipeline. The distance between the top of the jacking pipe and the upper pipeline is about 6.5m. During the jacking pipe construction, the thrust of the jacking machine is too large, and the soil in front of the excavation is squeezed, which may cause the surface to bulge and cause pipeline deformation; grouting under the pipeline may cause local bulge of the pipeline, which in turn causes pipeline deformation; due to soil water loss, soil disturbance, etc., the surface may sink, causing pipeline deformation;

[0261] Risk classification: Level I;

[0262] Countermeasures: In order to reduce the impact of pipe jacking construction on the upper pipeline, mechanical mud-water balance pipe jacking is used to ensure the stability of the face during the pipe jacking construction process, thereby minimizing the risk of deformation caused to the upper pipeline. During the jacking process, the direction of the tool pipe should be controlled, and the causes and development trends of the deviation should be analyzed based on the measurement results to determine the measures for correction. After the jacking is completed, the two sides should be reinforced with grouting. Multi-level waterproofing measures are adopted for pipe jacking. During the jacking process, the deformation and linearity of the steel pipe should be closely observed, and measurements and corrections should be made frequently. Strengthen the monitoring of the upper pipeline, establish an emergency linkage mechanism, make emergency plans and prepare emergency measures. During the pipe jacking process, synchronous grouting and secondary grouting should be done well. Grouting must follow the principle of "pressing first and then jacking, pressing as the jacking progresses, and timely grouting". The quality of grouting construction should be controlled mainly by pressure control and supplemented by grouting volume control. At the same time, technical parameters such as slag improvement, mud and water bin pressure and slag discharge volume should be strictly controlled during the excavation process to prevent the surface settlement from exceeding the limit.

[0263] Risk level after disposal: Guangzhou gas pipeline level III, Sinopec oil pipeline level II, Dapeng gas pipeline level II.

[0264] 5. Risk engineering categories: environmental risks;

[0265] Risk sources: upper ground;

[0266] Description of basic risk conditions: The soil covering the jacking tunnel of this project is mainly miscellaneous fill, silt, and silty clay. The burial depth of the jacking tunnel is 9.75m, and the span of a single jacking tunnel is 11.1m. The silt stratum is highly sensitive and easily disturbed during jacking construction. The surrounding soil has a weak constraint on the soil covering the tunnel. The surface is easily deformed during construction, resulting in uplift or settlement deformation of the surface.

[0267] Risk classification: Level II;

[0268] Countermeasures: During the actual advancement of the jacking pipe, according to the ground settlement monitoring data, the grouting amount (thixotropic mud), grouting location and grouting pressure are adjusted after analysis and judgment. During construction, thixotropic mud is added when necessary to effectively control the later settlement and conduct information-based dynamic construction management. When the jacking pipe is advancing under the pipeline, the parameters accumulated at the beginning are used for advancement. At the same time, full-time personnel should continuously monitor the settlement of the strata around the pipeline, monitor the deformation of the pipeline in time, and feedback the monitoring data to the jacking construction work surface in a timely and accurate manner, so that the central control room can make correct judgments based on the situation reflected on the ground and guide the timely optimization and adjustment of the jacking pipe advancement parameters. At the same time, according to the ground load, the earth pressure balance setting value is recalculated in time, and adjusted according to the ground heave value, so that the jacking pipe can be advanced quickly and evenly, the crossing time is shortened as much as possible, and over-excavation and under-excavation are prevented to reduce the disturbance of the soil, minimize the loss of the stratum, control the settlement within the minimum range, and meet the settlement requirements.

[0269] Risk level after disposal: Level III.

[0270] IV. Construction guarantee measures

[0271] 1. Guarantee measures for pipe jacking to pass through three large pipes

[0272] (1) Parameter collection and settlement control

[0273] Before construction, a detailed investigation was conducted on the three large pipes to determine the settlement and displacement of the pipelines within the impact range of the jacking construction, and compared them with the settlement during the down-penetration process to guide the construction.

[0274] (2) Simulation propulsion

[0275] A simulated advancement area was set up 10 meters before the city pipeline was passed under the city pipeline. Monitoring points for soil vertical and horizontal displacement at corresponding depths, groundwater level and water pressure monitoring points, and changes in deep monitoring points were monitored from beginning to end. The main purpose was to explore the influence of different construction parameters in the construction process on the disturbance of the strata within the depth range above the top pipe, and to explore the influence of different top pipe advancement speeds, mud and water tank pressures, and grouting processes (thixotropic mud) on the strata. The deformation of the strata and the set value of the mud and water tank pressure of the top pipe machine and the actual value of the top pipe excavation speed were accurately measured, and the principles of mathematical statistics were used to find the relationship between the above parameters. Through simulated advancement, the optimal construction parameters before the top pipe passed through the three large pipes were measured to guide the construction.

[0276] (3) Construction monitoring frequency

[0277] During the period of pipe jacking passing through the three large pipes, the monitoring frequency was increased. At the same time, various construction parameters were timely optimized and adjusted according to the monitoring data during the process of passing through the three large pipes.

[0278] (4) Settlement control

[0279] During the actual advancement of the pipe jacking, the grouting amount (thixotropic mud), grouting location and grouting pressure are adjusted after analysis and judgment based on the pipeline settlement situation.

[0280] During construction, thixotropic mud is injected when necessary to effectively control the later settlement and carry out information-based dynamic construction management.

[0281] At the same time, according to the ground load conditions, the mud balance setting value is recalculated in time and adjusted according to the ground heave value to enable the jacking pipe to advance quickly and evenly, shorten the crossing time as much as possible, prevent over-excavation and under-excavation, reduce the disturbance to the soil, minimize the loss of formation, control the settlement within the minimum range, and meet the settlement requirements.

[0282] Before the underpass construction, a comprehensive inspection of the pipe jacking machine is carried out to ensure that there are no mechanical failures during the crossing process, and daily maintenance is strengthened.

[0283] A. Strengthen construction management and pass through three large pipes at a uniform and continuous speed.

[0284] B. Control the internal pressure, adjust the viscosity and specific gravity of the thixotropic mud, avoid instability of the tunnel face, strengthen mud discharge management, and appropriately reduce the amount of mud discharge.

[0285] C. Ensure that the synchronous grouting system (thixotropic mud) works normally and increase the grouting volume appropriately.

[0286] D. Make preparations for surface grouting. When the settlement of urban roads exceeds the allowable value, carry out surface grouting immediately.

[0287] E. After the first crossing is completed, the original construction data is collected and the project department organizes a special discussion meeting to summarize the construction experience, optimize the construction parameters, and summarize the experience for the subsequent second underpass construction.

[0288] 2. Measures to ensure the use of pipe jacking machine

[0289] In order to ensure the normal use of the mud-water balanced rectangular pipe jacking machine, the following measures will be taken:

[0290] (1) Adjust the main parameters of the mud-water balanced rectangular pipe jacking machine and formulate and implement a scientific and reasonable maintenance and construction plan for the machine.

[0291] (2) Management of operating personnel

[0292] Drivers of mud-water balanced rectangular pipe jacking machines must undergo training and examination and pass the examination before they can take up their posts.

[0293] Maintenance personnel of mud-water balanced rectangular pipe jacking machines must be familiar with the working principle of the pipe jacking machine and the parts configuration of each subsystem of the machine.

[0294] Workers such as pipe joint assemblers and grouting workers must undergo professional training and strictly follow operating procedures.

[0295] Equipped with professional electrical engineers, who are required to be proficient in the electrical system of the pipe jacking machine.

[0296] (3) Precautions for promotion

[0297] Before pushing the pipe jacking machine forward, you must be familiar with the geological conditions ahead of the line. When encountering major changes in geological conditions, you must conduct exploration in advance and replace the cutter in time according to the soil quality.

[0298] When abnormalities are found in parameters such as thrust, cutter disc torque, propulsion speed, etc., stop the machine immediately for inspection and analyze the causes.

[0299] The position of each pipe section is measured, and an alarm is issued after comparing the manual measurement results with those of the pipe jacking machine.

[0300] Adjust propulsion parameters in time according to ground subsidence conditions.

[0301] When encountering difficulties ahead on the line, a promotion plan and emergency measures must be formulated first.

[0302] (4) Maintenance

[0303] Select the correct hydraulic oil, lubricating oil, sealing grease and other materials according to the operating manual.

[0304] The pipe jacking machine is routinely inspected and maintained every day.

[0305] Strictly follow the maintenance plan in the operation manual to carry out daily, weekly, monthly and semi-annual maintenance and inspection procedures for the machinery.

[0306] 3. Quality assurance measures for pipe jacking machines

[0307] (1) Strengthen the installation quality inspection of the rubber water stop strip at the hole opening; strengthen the tamping of the hole ring concrete to ensure that the hole ring concrete is dense; fill the gap between the pipe joint and the hole ring with hydraulic slurry to ensure the quality of the joint.

[0308] (2) The whole process of tunnel opening sealing shall be supervised by a dedicated quality inspector, and a dedicated person shall be arranged to conduct follow-up inspection on the tunnel opening sealing to ensure that the tunnel opening sealing is intact.

[0309] (3) Strengthen the design and verification of the backrest of the jacking pipe to ensure that the reaction wall has sufficient rigidity, strength and flatness. After the jacking pipe is assembled, the system and the whole machine are debugged and accepted by the supervision engineer. Only after approval can the construction be promoted.

[0310] (4) The surveying personnel shall accurately determine the centerline and elevation of the launching base. The installation accuracy of the launching platform and the backrest shall be strictly controlled. The launching base shall be welded firmly to ensure that the launching posture of the jacking pipe coincides with the designed line.

[0311] 4. Quality assurance measures for pipe jacking and synchronous grouting of pipe segments

[0312] (1) Strengthen the training of mud-water balanced rectangular pipe jacking construction personnel and improve the operating level of pipe jacking machine drivers.

[0313] (2) Strengthen construction measurement, adopt measures such as automatic measurement of mud-water balanced rectangular jacking pipes first, and manual measurement and verification at any time to ensure that the linearity of the tunnel and the position accuracy meet the requirements of the specifications.

[0314] (3) During the advancement of the mud-water balance rectangular pipe jacking machine, the mud quality is reasonably adjusted according to the different soil types and cover thickness, combined with ground monitoring information, and the advancement speed is reasonably controlled according to the relationship between thrust, advancement speed and excavation volume to ensure that the pressure in the bin is balanced with the pressure of the cement water bin on the excavation surface.

[0315] (4) Organize and advance the work in strict accordance with the design and construction specifications to improve the quality of pipe jacking.

[0316] (5) Develop detailed anti-friction grouting operation instructions, conduct detailed slurry ratio tests, and select appropriate grouting materials and slurry ratios.

[0317] (6) Develop detailed grouting construction design and process flow and grouting quality control procedures, strictly implement grouting, inspection, record and analysis as required, make P (grouting pressure)-Q (grouting volume)-t (time) curve in time, analyze the relationship between grouting speed and advancement speed, evaluate the grouting effect, and provide feedback to guide the next grouting.

[0318] (7) Establish a professional grouting operation team, with experienced grouting engineers responsible for on-site grouting technology and management.

[0319] (8) Based on the monitoring results of the deformation of the tunnel lining and the ground deformation, timely information feedback is provided to correct the grouting parameter design and construction methods, and timely resolve any problems found.

[0320] (9) Ensure the maintenance of grouting equipment and the supply of grouting materials, and regularly clean the grouting pipelines and equipment to ensure that the grouting operation is carried out smoothly and continuously without interruption.

[0321] (10) Improve the level of information management, provide timely feedback on quality information, analyze and process, and guide on-site construction.

[0322] 5. Pipe segment production quality assurance measures

[0323] (1) Production quality control

[0324] All indicators of pipe joint production should comply with the "Concrete Structure Engineering Construction Quality Acceptance Code (GB50204-2015)".

[0325] (2) Management of main production materials

[0326] Main materials such as concrete, rebar and steel are ensured to meet design requirements.

[0327] (3) Embedded parts

[0328] Grouting, grouting heads, embedded handhole parts, etc. are processed and produced according to the design drawings, and the supplier provides a factory certificate. Appearance inspection is carried out when entering the site, and unqualified products are handled according to requirements, and records are kept to ensure traceability.

[0329] 6. Tunnel axis quality assurance measures

[0330] (1) During the advancement, the direction and position of the pipe jacking machine should be grasped in time, and the posture of the pipe jacking machine should be strictly controlled to ensure that the deviation between the actual axis and the design axis is less than the requirement of ±50mm.

[0331] (2) During the pipe jacking process, measurements are carried out after each section is connected and installed. Frequent measurements and corrections are made to avoid error accumulation. The axis deviation should not exceed 4 mm at a time.

[0332] (3) Regularly measure the posture of the pipe jacking machine manually and correct any problems found in a timely manner.

[0333] 7. Quality assurance measures for pipe joint assembly

[0334] (1) Strictly install the pipe joint waterproofing materials according to the specifications and design requirements

[0335] The waterproof materials for pipe joints must have qualified certification materials and can only be implemented after inspection by the supervising engineer; the adhesive must be applied evenly and fully, the plywood and rubber rings must be pasted flatly and firmly, and the position must be accurate, without bulging, excess length or gaps; the groove must be cleaned before caulking, free of impurities and debris, the type, specification and quality of the caulking materials must meet the design and specification requirements, and can only be implemented after passing the supervisory inspection, and the machinery and equipment used must be in good condition.

[0336] (2) Reasonable selection of synchronous grouting parameters to ensure uniform stress on the pipe joints and achieve stability as soon as possible

[0337] Before grouting, carry out mix ratio test to select the best mix ratio, and adjust the mix ratio in time according to different geological conditions; strictly control the mixing quality of the slurry, ensure uniform mixing, sufficient time, and match the drag reduction mud triggering mud injection speed with the propulsion speed; control the extension and contraction of the propulsion cylinder and the grouting pressure, and control the assembly accuracy within the design requirements to prevent pipe segment displacement and misalignment.

[0338] 8. Measures for suspending excavation due to special circumstances during excavation

[0339] (1) After stopping the jacking, the pressure maintenance measures in the silo were taken according to the conventional measures, and the mud water pressure in the silo was set to 0.13-0.15mpa.

[0340] (2) Based on the pipeline detection inspection and measurement results, the steel plate is laid to the head position for laying.

[0341] (3) The slurry in the bin in front of the pipe jacking machine head adopts thixotropic mud to replace the circulating mud. By utilizing the characteristics of thixotropic mud, the machine is shut down to maintain the pressure, and the pressure is controlled at 0.13-0.15 MPa. At the same time, road surface monitoring is strengthened, and the mud pressure in the bin is inspected daily and personnel are assigned to be on duty.

[0342] (4) Strengthen the injection of thixotropic mud on the side wall of the jacking pipe, ensure that the thixotropic mud is injected on the outside of the pipe wall, and prevent the stagnation time from being too long, which will increase the friction and make it impossible to start the subsequent jacking construction. According to the previous experience of the construction unit, timely pressure maintenance and replenishment of thixotropic mud, the stagnation time is 1.5 months, and the jacking distance is 26m.

[0343] (5) When conditions permit, promptly enclose the nose of the aircraft and block the nose of the aircraft to prevent vehicles and pedestrians from passing by.

[0344] (6) Establish an emergency liaison group with relevant authority units to facilitate emergency management and control work.

[0345] (7) Emergency rescue operations can be carried out according to the pipe jacking emergency plan.

[0346] Civilized construction guarantee measures, environmental protection measures, summer construction and guarantee measures, rainy season (heavy rain) construction and guarantee measures, typhoon weather construction and guarantee measures.

[0347] V. Emergency Plan

[0348] 1. Emergency Response Procedures

[0349] 1) Early warning

[0350] (1) Early warning system

[0351] Relying on the risk management monitoring system established by Guangzhou Metro Group Co., Ltd. in all participating units, combined with the construction characteristics of this project, we will strengthen the monitoring of major hazardous sources of this project and the impact on the surrounding environment of the project, take preventive control measures for factors that may cause accidents, and achieve early detection, early reporting, and early disposal.

[0352] When the project site personnel or monitoring personnel find that there is a dangerous situation, the various parameters of the dangerous source exceed the normal value limit or an accident has occurred, they must report it to the emergency rescue leadership group quickly and promptly, and launch preliminary rescue operations at the same time.

[0353] After receiving information that may lead to a production safety accident, the emergency rescue leading group will, on the one hand, promptly study and determine the response plan according to the emergency plan, and notify the relevant departments and operators of the project department to take corresponding actions to prevent the accident, and report to the company, the supervision unit, and the owner unit. On the other hand, it will accurately determine the degree of harm and the level of the accident, and quickly and promptly inform the relevant personnel to evacuate the dangerous situation.

[0354] (2) Warning level

[0355] Based on monitoring and forecasting analysis, the warning levels are divided from low to high into yellow, orange and red according to factors such as the degree of harm, development trend and urgency of various events that may cause sudden accidents.

[0356] a. Yellow Warning

[0357] A blue alert may be issued when one of the following conditions is met:

[0358] ① It is expected that a safety accident of level 3 or above will occur, the accident is imminent, and the situation may escalate;

[0359] ② When relevant departments issue a yellow warning for severe weather such as strong winds, heavy snow, heavy rain, and high temperatures;

[0360] ③ It is verified that the change rate of monitoring data at more than 10 monitoring points reaches 80% of the control value of the change rate of the point; or the cumulative value of monitoring data at more than 5 (including 5) and less than 10 monitoring points reaches 80% of the control value of the point and the monitoring data does not converge; or there are less than 5 monitoring points whose change rate or cumulative value reaches the control value of the point; or more than 2 monitoring types appear in a certain monitoring section at the same time at a change rate of 80% of the control value;

[0361] ④ The on-site inspection found that there were major safety hazards in the construction, but there were no obvious signs of damage such as cracks in the project structure itself and the surrounding environment, and no timely rectification was made;

[0362] ⑤ One of the “dual control” indicators of construction project monitoring data reaches the control indicator value between 80% and 100%.

[0363] ⑥During the on-site inspection, it was discovered that the project was in an unsafe state with slight safety risks.

[0364] b. Orange warning

[0365] An orange alert may be issued when any of the following conditions are met:

[0366] ① It is expected that a safety accident of level 2 or above will occur, the accident is imminent, and the situation may escalate;

[0367] ② When relevant departments issue orange warnings for severe weather such as strong winds, heavy snow, heavy rain, and high temperatures;

[0368] ③ It is verified that the change rate of monitoring data at more than 10 monitoring points reaches 80% of the control value of the change rate of the point and the monitoring data does not converge; or the change rate or cumulative value of monitoring data at more than 5 (including 5) monitoring points reaches the control value and the monitoring data does not converge; or more than 2 monitoring types appear in a certain monitoring section at the same time, and the cumulative value reaches 80% of the control value;

[0369] ④ The on-site inspection found obvious safety hazards in the construction, such as a small number of cracks in the project structure and the surrounding environment, excessive deformation of a small number of components, local buckling of some components and other signs of damage, and no timely rectification;

[0370] ⑤ The "dual control" indicators of the construction project monitoring data both reach between 80% and 100% of the control indicator value, or one of the "dual control" indicators exceeds the control indicator value.

[0371] ⑥During the on-site inspection, it was discovered that the project was in an unsafe state with serious safety risks.

[0372] c. Red Alert

[0373] A red alert may be issued when any of the following conditions are met:

[0374] ① It is expected that a safety accident of level 1 or above will occur, the accident is imminent, and the situation may escalate;

[0375] ② When relevant departments issue a red warning for severe weather such as strong winds, heavy snow, heavy rain, and high temperatures;

[0376] ③ It is verified that the change rate of monitoring data at more than 10 monitoring points reaches or the cumulative value reaches the control value of the point and the monitoring data does not converge; or more than two types of monitoring types appear in a certain monitoring section at the same time at a change rate and cumulative value reaching the control value;

[0377] ④ The on-site inspection revealed serious safety hazards in the construction process. Large cracks appeared in the structure and the surrounding environment. Some components were deformed too much and showed a trend of further development. Some components showed local buckling and other obvious signs of damage.

[0378] ⑤The “dual control” indicators of the construction project monitoring data all exceed the control indicator values.

[0379] ⑥During the on-site inspection, it was discovered that the project was in an unsafe state with serious safety risks.

[0380] (3) Issuance and cancellation of warnings

[0381] The surveying team is responsible for reporting monitoring and early warning information; the safety supervision department is responsible for reporting severe weather warning information; the safety supervision department is responsible for reporting safety accident warning information, and the emergency rescue leading group office shall promptly summarize and report the warning information.

[0382] The emergency rescue leading group shall promptly analyze the hidden dangers of emergencies and various types of information, and assess the possibility of emergencies and their possible impacts; for those that may cause emergencies, after judging based on the classification of warning levels, the emergency rescue leading group shall discuss and decide and issue an event warning of the corresponding level, and announce that the relevant area has entered the warning period.

[0383] The project department's emergency rescue leadership group is responsible for issuing warning information and raising or lowering the warning level accordingly based on the development of the situation.

[0384] The early warning information includes the type of production safety accident, warning level, start time, possible impact range, warning matters, measures to be taken, etc.

[0385] When the conditions that may lead to an accident are eliminated, the emergency rescue leadership team of the project department will publicize and lift the warning. The release, adjustment and lifting of warning information can be carried out through communication, information network and other means.

[0386] (4) Early warning response

[0387] After the early warning information is released, the project emergency rescue leading group, safety supervision department, engineering department, material department, surveying team and other project departments and teams enter the emergency response work state.

[0388] Yellow Warning Response

[0389] Further strengthen leadership and on-site inspections, and promptly handle and report problems when they are discovered. There will be people on duty at the construction site day and night, and communication links will be kept open at all times. Construction site managers and safety officers will be on duty, strictly stop and correct illegal construction, eliminate safety hazards as soon as possible, and pay close attention to natural conditions, monitoring data, and various changes at the construction site, and make all preparations for emergency rescue.

[0390] ② Orange warning response

[0391] On the basis of the yellow warning response, all departments pay close attention to the rectification of hidden dangers and natural disasters. The heads of all departments should keep abreast of the situation, eliminate potential safety hazards, and promptly report the warning information and analysis and evaluation results to the project department's emergency rescue leadership team. If necessary, adjust the construction period or stop on-site construction. The project department's emergency team is on standby at any time and will immediately carry out rescue after receiving the accident report.

[0392] ③ Red Alert Response

[0393] On the basis of the orange warning response, all departments should pay close attention to the dynamics of major hazards and natural disasters, and the leaders on duty should take the initiative to understand and grasp the situation, strengthen the density of duty and monitoring, and stop construction work when necessary.

[0394] When a yellow, orange or red alert is issued, in addition to the specific measures taken during a blue alert, one or more of the following measures should be taken in light of the characteristics of the impending emergency and the possible harm it may cause:

[0395] a. Members of the project department's emergency rescue team and members of each emergency rescue team enter standby status;

[0396] b. Mobilize the materials, equipment and tools needed for emergency rescue and ensure that they are in good condition and can be put into normal use at any time;

[0397] c. Strengthen security protection and safety inspections at key and important locations;

[0398] d. In accordance with the instructions of the owner or the government, timely issue warnings to neighboring units and residents who may be harmed about possible harm from the emergency, and publicize common sense on how to avoid and reduce harm;

[0399] e. Other necessary preventive and protective measures required by laws, regulations, rules and regulations.

[0400] It should be noted that, for the aforementioned method embodiments, for the sake of simplicity, they are all expressed as a series of action combinations, but those skilled in the art should be aware that the present application is not limited by the described order of actions, because according to the present application, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required by the present application.

[0401] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0402] In the several embodiments provided in the present application, it should be understood that the disclosed device can be implemented in other ways. For example, the device embodiments described above are only schematic, such as the division of the above-mentioned units, which is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, and the indirect coupling or communication connection of devices or units can be electrical or other forms.

[0403] The units described above as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0404] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0405] If the above-mentioned integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable memory. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a memory, including a number of instructions to enable a computer device (which can be a personal computer, server or network device, etc.) to perform all or part of the steps of the above-mentioned methods of each embodiment of the present application. The aforementioned memory includes: U disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), mobile hard disk, disk or CD-ROM and other media that can store program codes.

[0406] A person skilled in the art can understand that all or part of the steps in the various methods of the above embodiments can be completed by instructing related hardware through a program, and the program can be stored in a computer-readable memory, and the memory can include: a flash drive, a read-only memory (English: Read-Only Memory, abbreviated as: ROM), a random access memory (English: Random Access Memory, abbreviated as: RAM), a magnetic disk or an optical disk, etc.

[0407] The embodiments of the present application are introduced in detail above. Specific examples are used in this article to illustrate the principles and implementation methods of the present application. The description of the above embodiments is only used to help understand the method and core idea of ​​the present application. At the same time, for general technical personnel in this field, according to the idea of ​​the present application, there will be changes in the specific implementation method and application scope. In summary, the content of this specification should not be understood as a limitation on the present application.

[0408] Although the present invention is disclosed as above, the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions without departing from the spirit and scope of the present invention, and can make various changes and modifications, including the combination of the above-mentioned different functions and implementation steps, including software and hardware implementation methods, all of which are within the scope of protection of the present invention.

Claims

1. A decision-making system for rail transit engineering data analysis, characterized in that: include: Cloud servers, IoT smart sensors and IoT servers; among them, The cloud server is configured as follows: Acquiring first data related to the rail transit project; Constructing a first virtual digital twin model including all rail transit engineering data according to the first data; Applying big data technology, summarizing and analyzing the first data, obtaining first construction risk data and first key node data, and formulating corresponding first response measures; Performing scheme simulation and collision detection according to the first virtual digital twin model, the first construction risk data, the first key node data, and the first countermeasures to obtain a plurality of first construction schemes; Combining the first data and the first construction plan and using a deep learning algorithm to train a first construction model including a construction guidance model, a construction progress prediction model, a quality risk model, a safety accident risk model, and a preventive measures planning model; According to the first construction model, a plurality of the IoT smart sensors are installed at the construction site to collect second data during the construction process; Establish a knowledge graph and a decision tree based on the first construction model and the second data, develop a construction decision support system based on a rule engine and artificial intelligence, and provide decision suggestions; The step of applying big data technology to summarize and analyze the first data to obtain first construction risk data and first key node data, and formulating corresponding first response measures includes: Cleaning, denoising, and integrating the first data to form a structured first data set, and using the first data set as a basic data source for big data analysis; Using statistics and data mining techniques, analyzing the first data set to identify key influencing factors and establishing a first model reflecting project progress and risks; Based on the first model, a risk identification algorithm is developed to dynamically identify potential problems and risks according to the project schedule and quality requirements, thereby obtaining the first construction risk data; In combination with the project schedule, identify the key nodes with tight construction period and resources and the important process nodes that have the greatest impact on the overall project schedule, and obtain the first key node data from the key nodes and the important process nodes; According to the first construction risk data and the first key node data, using the expert knowledge base, a corresponding emergency plan and resource scheduling solution are provided as the first response measure; The step of using statistics and data mining techniques to analyze the first data set to identify key influencing factors and establish a first model reflecting project progress and risks includes: extracting a first characteristic variable related to project progress and risk from the first data set through domain knowledge; Using statistical methods to identify the correlation between the first characteristic variables, and deleting redundant variables to obtain second characteristic variables; Using a machine learning method to analyze the importance of each of the second characteristic variables to determine the key influencing factors; Combining the second characteristic variable and the key influencing factor, using a regression algorithm to train a first project schedule prediction model; Combining the second characteristic variable and the key influencing factor, using a classification algorithm to train a first risk classification model; fusing the first project schedule prediction model and the first risk classification model into the first model; The step of installing a plurality of IoT smart sensors at the construction site according to the first construction model to collect second data during the construction process includes: Determining the data input requirements of the model for sensor data according to the first construction model; Determine the types of sensors to be installed and the sensor coverage of various sensors according to the data input requirements; Calculating the required number of sensors according to the construction site area, distribution of important areas, and coverage of the sensors; Formulate a first distribution plan of sensors according to the data requirements of different regions of the first construction model, the sensor types, and the number of sensors; Obtaining power supply data, network wiring data, and network coverage data of the construction area; According to the power supply data, the network wiring data, and the network coverage data, the first virtual digital twin model is used to simulate, verify, and modify the first distribution plan to ensure that the data input requirements of the first construction model can be met; According to the first distribution scheme, a plurality of the IoT smart sensors are installed to collect the second data; During the construction process, the sensor layout is dynamically adjusted according to changes in the data input requirements of the first construction model.

2. A decision analysis method for rail transit engineering data analysis, characterized in that: include: Acquiring first data related to the rail transit project; Constructing a first virtual digital twin model including all rail transit engineering data according to the first data; Applying big data technology, summarizing and analyzing the first data, obtaining first construction risk data and first key node data, and formulating corresponding first response measures; Performing scheme simulation and collision detection according to the first virtual digital twin model, the first construction risk data, the first key node data, and the first countermeasures to obtain a plurality of first construction schemes; Combining the first data and the first construction plan and using a deep learning algorithm to train a first construction model including a construction guidance model, a construction progress prediction model, a quality risk model, a safety accident risk model, and a preventive measures planning model; According to the first construction model, a plurality of IoT smart sensors are installed at the construction site to collect second data during the construction process; Establish a knowledge graph and a decision tree based on the first construction model and the second data, develop a construction decision support system based on a rule engine and artificial intelligence, and provide decision suggestions; The step of applying big data technology to summarize and analyze the first data to obtain first construction risk data and first key node data, and formulating corresponding first response measures includes: Cleaning, denoising, and integrating the first data to form a structured first data set, and using the first data set as a basic data source for big data analysis; Using statistics and data mining techniques, analyzing the first data set to identify key influencing factors and establishing a first model reflecting project progress and risks; Based on the first model, a risk identification algorithm is developed to dynamically identify potential problems and risks according to the project schedule and quality requirements, thereby obtaining the first construction risk data; In combination with the project schedule, identify the key nodes with tight construction period and resources and the important process nodes that have the greatest impact on the overall project schedule, and obtain the first key node data from the key nodes and the important process nodes; According to the first construction risk data and the first key node data, using the expert knowledge base, a corresponding emergency plan and resource scheduling solution are provided as the first response measure; The step of using statistics and data mining techniques to analyze the first data set to identify key influencing factors and establish a first model reflecting project progress and risks includes: extracting a first characteristic variable related to project progress and risk from the first data set through domain knowledge; Using statistical methods to identify the correlation between the first characteristic variables, and deleting redundant variables to obtain second characteristic variables; Using a machine learning method to analyze the importance of each of the second characteristic variables to determine the key influencing factors; Combining the second characteristic variable and the key influencing factor, using a regression algorithm to train a first project schedule prediction model; Combining the second characteristic variable and the key influencing factor, using a classification algorithm to train a first risk classification model; fusing the first project schedule prediction model and the first risk classification model into the first model; The step of installing a plurality of IoT smart sensors at the construction site according to the first construction model to collect second data during the construction process includes: Determining the data input requirements of the model for sensor data according to the first construction model; Determine the types of sensors to be installed and the sensor coverage of various sensors according to the data input requirements; Calculating the required number of sensors according to the construction site area, distribution of important areas, and coverage of the sensors; Formulate a first distribution plan of sensors according to the data requirements of different regions of the first construction model, the sensor types, and the number of sensors; Obtaining power supply data, network wiring data, and network coverage data of the construction area; According to the power supply data, the network wiring data, and the network coverage data, the first virtual digital twin model is used to simulate, verify, and modify the first distribution plan to ensure that the data input requirements of the first construction model can be met; According to the first distribution scheme, a plurality of the IoT smart sensors are installed to collect the second data; During the construction process, the sensor layout is dynamically adjusted according to changes in the data input requirements of the first construction model.

3. The decision analysis method for rail transit engineering data analysis according to claim 2 is characterized in that: The step of developing a risk identification algorithm based on the first model, dynamically identifying potential problems and risks according to the project schedule and quality requirements, and obtaining the first construction risk data includes: Collect historical project risk cases and build a standardized first risk database; Acquire real-time progress and real-time quality data of the project, and match the first risk type that is expected to occur according to the trigger conditions in the first risk library; Using the first risk classification model, evaluating the matched first risk type to obtain a first risk evaluation result; Using the first project progress prediction model, predict the subsequent progress and completion time of the project to obtain a first progress prediction result; combining the first risk assessment result with the first progress prediction result to dynamically identify the first construction risk data that may affect the project progress; The identified first construction risk data is saved in the form of a structured data set.

4. The decision analysis method for rail transit engineering data analysis according to claim 3 is characterized in that: The step of identifying the key nodes with tight construction period and resources and the important process nodes that have the greatest impact on the overall project schedule in combination with the project schedule, and acquiring the first key node data from the key nodes and the important process nodes includes: Extracting a pipe jacking construction schedule from the project schedule; The data of key points in the pipe jacking construction schedule, including geological condition data, tunnel burial depth data, tunnel spacing, and starting and arrival engineering data of the pipe jacking, are identified as the first key node data.

5. The decision analysis method for rail transit engineering data analysis according to claim 4 is characterized in that: The step of providing a corresponding emergency plan and resource scheduling solution as the first response measure using an expert knowledge base based on the first construction risk data and the first key node data includes: Collect emergency plan templates and resource scheduling plan examples in the construction field to form a structured emergency plan knowledge base; Determining a matching first emergency plan template in the emergency plan knowledge base according to the first construction risk data; Determine a matching first resource scheduling scheme according to the first key node data and resource status data; Evaluate the first emergency plan template and the first resource scheduling scheme, and adjust and improve the first emergency plan template and the first resource scheduling scheme according to the evaluation results; Simulating the execution effect of the first emergency plan template and the first resource scheduling solution in the first virtual digital twin system; The first emergency plan template and the first resource scheduling plan are adjusted according to the execution effect to obtain the first response measure.

6. The decision analysis method for rail transit engineering data analysis according to claim 5, characterized in that: The step of performing scheme simulation and collision detection according to the first virtual digital twin model, the first construction risk data, the first key node data and the first countermeasures to obtain a plurality of first construction schemes includes: Importing the first construction risk data and the first key node data into the first virtual digital twin model for association and annotation; Setting simulation parameters of resources, environment, and construction process in the first virtual digital twin model, and defining simulation conditions of different construction schemes; Simulate the execution of different construction plans and calculate the time schedule, resource utilization rate and risk elimination effect of each construction plan; During the simulation, a collision detection algorithm is used to check whether different construction plans conflict in time and space; Based on the simulation and collision detection results, multiple feasible construction plans that meet the time, resource and risk constraints are generated, and quantitative and qualitative evaluation and analysis are performed on different feasible construction plans to determine multiple first construction plans.

7. The decision analysis method for rail transit engineering data analysis according to claim 6, characterized in that: The step of combining the first data and the first construction plan and using a deep learning algorithm to train a first construction model including a construction guidance model, a construction progress prediction model, a quality risk model, a safety accident risk model, and a preventive measures planning model comprises: Extracting first construction data including BIM data, sensor data, document data, video data, and audio data from the first data; Extracting data about construction instructions from the first construction data, and building the construction instruction model based on the CNN neural network in combination with the first construction plan; Extracting data on project progress and resource utilization from the first construction data, and building the LSTM-based predicted construction progress model in combination with the first construction plan; Extracting usage quality inspection data and quality requirement data from the first construction data, and building the quality risk model based on a convolutional neural network in combination with the first construction plan; Extracting safety record data and safety specification data from the first construction data, and training the multi-modal safety accident risk model in combination with the first construction plan; Extracting historical preventive measures data and preset preventive requirements data from the first construction data, and building the preventive measures planning model based on the knowledge graph in combination with the first construction plan; The first construction model driven by deep learning is constructed by integrating the construction guidance model, the predicted construction progress model, the quality risk model, the safety accident risk model and the preventive measures planning model.

Images